From Moray.Henderson at ict-software.org Tue May 3 09:09:57 2011 From: Moray.Henderson at ict-software.org (Moray Henderson) Date: Tue, 3 May 2011 10:09:57 +0100 Subject: Use entire disk as PV. In-Reply-To: References: Message-ID: <000501cc0971$e1033520$a3099f60$@Henderson@ict-software.org> From: Speagle, Andy [mailto:andy.speagle at wichita.edu] I'm trying to setup the partitioning section of my kickstarts in such a way that rather than partitioning a disk and using /dev/sdX1 as the PV for my root VG, that I can instead use the entire disk. The reason for doing this would be to make PV resizing a bit easier for virtual machines. Otherwise, I must muck around with the partition table and blah blah . not a show-stopper, but potentially dangerous for junior admins. I suspect that the functionality just isn't there. but does anyone know the magic syntax to get it to use the entire drive (aka. /dev/sdb) as a PV during kickstart? Why do you need to resize PVs for virtual machines? You allocate the extra space to your VM as another virtual drive or partition, then using LVM create another Physical Volume for that drive, add that Physical Volume to your Volume Group and Logical Volume, and extend the filesystem. No mucking around with an existing partition table. I would be confused if I encountered a disk without a partition table - I'd probably assume it wasn't formatted at all. Better to train your junior admins in standard practice? Moray. "To err is human; to purr, feline." -------------- next part -------------- An HTML attachment was scrubbed... URL: From feng.tan at oracle.com Tue May 3 09:26:13 2011 From: feng.tan at oracle.com (feng.tan at oracle.com) Date: Tue, 3 May 2011 02:26:13 -0700 (PDT) Subject: Auto Reply: RE: Use entire disk as PV. Message-ID: ???????????????? Dear Sender, >From May 3 ~ May 6, I am in Vacation, in this period I can not check email frequently. For Lab/Sending notification/Import/Export issue, please contact Carson Zhu(mailto:carson.zhu at oracle.com) or my manager Solon Zhang(mailto:solon.zhang at oracle.com). If there is any urgent case, you can reach me at 86-135-010-88029 Thanks, Tan Feng From feng.tan at oracle.com Tue May 3 09:35:16 2011 From: feng.tan at oracle.com (feng.tan at oracle.com) Date: Tue, 3 May 2011 02:35:16 -0700 (PDT) Subject: Auto Reply: Auto Reply: RE: Use entire disk as PV. Message-ID: <13da86a6-db4b-4fcf-860a-979190105a38@default> ???????????????? Dear Sender, >From May 3 ~ May 6, I am in Vacation, in this period I can not check email frequently. For Lab/Sending notification/Import/Export issue, please contact Carson Zhu(mailto:carson.zhu at oracle.com) or my manager Solon Zhang(mailto:solon.zhang at oracle.com). If there is any urgent case, you can reach me at 86-135-010-88029 Thanks, Tan Feng From CallahanT at tessco.com Tue May 3 10:45:19 2011 From: CallahanT at tessco.com (Callahan, Tom) Date: Tue, 3 May 2011 06:45:19 -0400 Subject: Use entire disk as PV. In-Reply-To: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> Message-ID: Most storage arrays can grow existing disks, which doesn't work as nicely if the disk has a partition. Making a PV out of the entire disk allows much easier PV/VG/LV expansion, without mucking with a partition table. As for the training piece, I'd expect an admin would do some verification a disk is not in use before blindly assuming it's not in use, and overwriting data. There is no standard practice with how to create a PV, it depends on the site. On May 3, 2011, at 5:30 AM, "Moray Henderson" wrote: > From: Speagle, Andy [mailto:andy.speagle at wichita.edu] > > > I?m trying to setup the partitioning section of my kickstarts in such a way that rather than partitioning a disk and using /dev/sdX1 as the PV for my root VG, that I can instead use the entire disk. > > > > The reason for doing this would be to make PV resizing a bit easier for virtual machines. Otherwise, I must muck around with the partition table and blah blah ? not a show-stopper, but potentially dangerous for junior admins. > > > > I suspect that the functionality just isn?t there? but does anyone know the magic syntax to get it to use the entire drive (aka. /dev/sdb) as a PV during kickstart? > > > > Why do you need to resize PVs for virtual machines? You allocate the extra space to your VM as another virtual drive or partition, then using LVM create another Physical Volume for that drive, add that Physical Volume to your Volume Group and Logical Volume, and extend the filesystem. No mucking around with an existing partition table. > > > > I would be confused if I encountered a disk without a partition table ? I?d probably assume it wasn?t formatted at all. Better to train your junior admins in standard practice? > > > > > > Moray. > > ?To err is human; to purr, feline.? > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From feng.tan at oracle.com Tue May 3 10:56:17 2011 From: feng.tan at oracle.com (feng.tan at oracle.com) Date: Tue, 3 May 2011 03:56:17 -0700 (PDT) Subject: Auto Reply: Re: Use entire disk as PV. Message-ID: <6e56518b-4494-43f0-a26a-7244c01f4e9e@default> ???????????????? Dear Sender, >From May 3 ~ May 6, I am in Vacation, in this period I can not check email frequently. For Lab/Sending notification/Import/Export issue, please contact Carson Zhu(mailto:carson.zhu at oracle.com) or my manager Solon Zhang(mailto:solon.zhang at oracle.com). If there is any urgent case, you can reach me at 86-135-010-88029 Thanks, Tan Feng From andy.speagle at wichita.edu Tue May 3 12:16:58 2011 From: andy.speagle at wichita.edu (Speagle, Andy) Date: Tue, 3 May 2011 07:16:58 -0500 Subject: Use entire disk as PV. In-Reply-To: References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> Message-ID: That is precisely my rationale. We?re using ESXi; resizing a disk is a trivial event that makes more sense for our environment. I also agree with Tom on the training piece. Woe be unto the admin who blindly uses a disk without any investigation and/or verification. Training junior admins to understand all of the possible configurations and to look before they leap are key parts of said training? -Andy From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: Tuesday, May 03, 2011 5:45 AM To: Discussion list about Kickstart Cc: Discussion list about Kickstart Subject: Re: Use entire disk as PV. Most storage arrays can grow existing disks, which doesn't work as nicely if the disk has a partition. Making a PV out of the entire disk allows much easier PV/VG/LV expansion, without mucking with a partition table. As for the training piece, I'd expect an admin would do some verification a disk is not in use before blindly assuming it's not in use, and overwriting data. There is no standard practice with how to create a PV, it depends on the site. On May 3, 2011, at 5:30 AM, "Moray Henderson" > wrote: From: Speagle, Andy [mailto:andy.speagle at wichita.edu] I?m trying to setup the partitioning section of my kickstarts in such a way that rather than partitioning a disk and using /dev/sdX1 as the PV for my root VG, that I can instead use the entire disk. The reason for doing this would be to make PV resizing a bit easier for virtual machines. Otherwise, I must muck around with the partition table and blah blah ? not a show-stopper, but potentially dangerous for junior admins. I suspect that the functionality just isn?t there? but does anyone know the magic syntax to get it to use the entire drive (aka. /dev/sdb) as a PV during kickstart? Why do you need to resize PVs for virtual machines? You allocate the extra space to your VM as another virtual drive or partition, then using LVM create another Physical Volume for that drive, add that Physical Volume to your Volume Group and Logical Volume, and extend the filesystem. No mucking around with an existing partition table. I would be confused if I encountered a disk without a partition table ? I?d probably assume it wasn?t formatted at all. Better to train your junior admins in standard practice? Moray. ?To err is human; to purr, feline.? _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From feng.tan at oracle.com Tue May 3 12:29:13 2011 From: feng.tan at oracle.com (feng.tan at oracle.com) Date: Tue, 3 May 2011 05:29:13 -0700 (PDT) Subject: Auto Reply: RE: Use entire disk as PV. Message-ID: <758cd6c2-83d5-472f-a3ca-d92f53064fc6@default> ???????????????? Dear Sender, >From May 3 ~ May 6, I am in Vacation, in this period I can not check email frequently. For Lab/Sending notification/Import/Export issue, please contact Carson Zhu(mailto:carson.zhu at oracle.com) or my manager Solon Zhang(mailto:solon.zhang at oracle.com). If there is any urgent case, you can reach me at 86-135-010-88029 Thanks, Tan Feng From beyonddc.storage at gmail.com Tue May 3 21:34:43 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Tue, 3 May 2011 17:34:43 -0400 Subject: Monitoring kickstart install status Message-ID: Hi All, I am hoping to run a python script in the background during %pre to monitor the kickstart install status. My script will periodically send status to a server somewhere within LAN and report the current installation status. Running a python script in %pre and sending status to a remote server is the easy part but I am not sure what to monitor to retrieve the kickstart install status. One thought I have is to monitor the /mnt/sysimage/root/install.log and compare the number of packages installed and the number of packages specified in the ks.cfg and get a rough estimate of the kickstart install status. I am not sure if this is a correct way to do or there's a better to do so. Please let me know if you've any suggestion or thought. Thanks! - David -------------- next part -------------- An HTML attachment was scrubbed... URL: From feng.tan at oracle.com Tue May 3 22:11:57 2011 From: feng.tan at oracle.com (feng.tan at oracle.com) Date: Tue, 3 May 2011 15:11:57 -0700 (PDT) Subject: Auto Reply: Monitoring kickstart install status Message-ID: <11ac2b63-28e8-4daa-9460-c9d053f190af@default> ???????????????? Dear Sender, >From May 3 ~ May 6, I am in Vacation, in this period I can not check email frequently. For Lab/Sending notification/Import/Export issue, please contact Carson Zhu(mailto:carson.zhu at oracle.com) or my manager Solon Zhang(mailto:solon.zhang at oracle.com). If there is any urgent case, you can reach me at 86-135-010-88029 Thanks, Tan Feng From root at nachtmaus.us Tue May 3 23:59:07 2011 From: root at nachtmaus.us (root at nachtmaus.us) Date: Tue, 3 May 2011 23:59:07 +0000 Subject: Monitoring kickstart install status In-Reply-To: References: Message-ID: <2077006858-1304467149-cardhu_decombobulator_blackberry.rim.net-856587006-@bda614.bisx.prod.on.blackberry> Remember, in the ks, you can specify group-installs, and dependancies should be resolved, so the number of packages in the ks will typically tend to be low. Instead, look for specific marker packages to be installed, such as kernel, or your standard userspace, or service-specific packages (apache, if it is a web-server; dovecot, if it is a mail/groupware server; local application package, etc) and send a syslog message/snmp trap/MQ message when you see one of these markers. -DTK Sent via BlackBerry from T-Mobile -----Original Message----- From: Chun Tat David Chu Sender: kickstart-list-bounces at redhat.com Date: Tue, 3 May 2011 17:34:43 To: Reply-To: Discussion list about Kickstart Subject: Monitoring kickstart install status _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From hbrown at divms.uiowa.edu Wed May 4 00:01:50 2011 From: hbrown at divms.uiowa.edu (Hugh Brown) Date: Tue, 03 May 2011 19:01:50 -0500 Subject: Monitoring kickstart install status In-Reply-To: References: Message-ID: <4DC0976E.9070800@divms.uiowa.edu> On 05/03/2011 04:34 PM, Chun Tat David Chu wrote: > Hi All, > > I am hoping to run a python script in the background during %pre to monitor > the kickstart install status. My script will periodically send status to a > server somewhere within LAN and report the current installation status. > > Running a python script in %pre and sending status to a remote server is the > easy part but I am not sure what to monitor to retrieve the kickstart > install status. > > One thought I have is to monitor the /mnt/sysimage/root/install.log and > compare the number of packages installed and the number of packages > specified in the ks.cfg and get a rough estimate of the kickstart install > status. > > I am not sure if this is a correct way to do or there's a better to do so. > > Please let me know if you've any suggestion or thought. > > Thanks! > > - David My memory says that the install.log is actually in /tmp/install.log. After the installation is finished it is copied to /mnt/sysimage/root/install.log. For fedora systems, I enabled sshd and require a password: sshpw --username=root XXXXXXXX --plaintext If I want to check status, I ssh in and look. If I don't, there's a script that will email me when it has finished. On RH5 systems, I enabled vnc and would use vncviewer to connect and check progress. Otherwise, watching the number of packages installed in install.log works. Hugh From feng.tan at oracle.com Wed May 4 00:09:52 2011 From: feng.tan at oracle.com (feng.tan at oracle.com) Date: Tue, 3 May 2011 17:09:52 -0700 (PDT) Subject: Auto Reply: Re: Monitoring kickstart install status Message-ID: <30cd6ceb-3410-4795-bcf8-31c4b6b8a24f@default> ???????????????? Dear Sender, >From May 3 ~ May 6, I am in Vacation, in this period I can not check email frequently. For Lab/Sending notification/Import/Export issue, please contact Carson Zhu(mailto:carson.zhu at oracle.com) or my manager Solon Zhang(mailto:solon.zhang at oracle.com). If there is any urgent case, you can reach me at 86-135-010-88029 Thanks, Tan Feng From feng.tan at oracle.com Wed May 4 01:11:19 2011 From: feng.tan at oracle.com (feng.tan at oracle.com) Date: Tue, 3 May 2011 18:11:19 -0700 (PDT) Subject: Auto Reply: Re: Monitoring kickstart install status Message-ID: <153f5cf6-c5fd-44fc-9217-a7f5c01bc456@default> ???????????????? Dear Sender, >From May 3 ~ May 6, I am in Vacation, in this period I can not check email frequently. For Lab/Sending notification/Import/Export issue, please contact Carson Zhu(mailto:carson.zhu at oracle.com) or my manager Solon Zhang(mailto:solon.zhang at oracle.com). If there is any urgent case, you can reach me at 86-135-010-88029 Thanks, Tan Feng From member at linkedin.com Wed May 4 06:07:52 2011 From: member at linkedin.com (Nauman yousuf via LinkedIn) Date: Wed, 4 May 2011 06:07:52 +0000 (UTC) Subject: Invitation to connect on LinkedIn Message-ID: <2007580608.579276.1304489272108.JavaMail.app@ela4-bed83.prod> LinkedIn ------------Nauman yousuf requested to add you as a connection on LinkedIn: ------------------------------------------ prasad, I'd like to add you to my professional network on LinkedIn. - Nauman Accept invitation from Nauman yousuf http://www.linkedin.com/e/gqo4xa-gn9v8ca1-1k/uDOOeOoBcDnu7cRsgtmHGao4wzLdPX8f936x8gS/blk/I104627508_25/1BpC5vrmRLoRZcjkkZt5YCpnlOt3RApnhMpmdzgmhxrSNBszYRcBYUc3kTczoQc359bS93dQRUjmpMbPwMc3sOejgUc38LrCBxbOYWrSlI/EML_comm_afe/ View invitation from Nauman yousuf http://www.linkedin.com/e/gqo4xa-gn9v8ca1-1k/uDOOeOoBcDnu7cRsgtmHGao4wzLdPX8f936x8gS/blk/I104627508_25/3kOnPwMdjsOdzgMckALqnpPbOYWrSlI/svi/ ------------------------------------------ DID YOU KNOW LinkedIn can help you find the right service providers using recommendations from your trusted network? Using LinkedIn Services, you can take the risky guesswork out of selecting service providers by reading the recommendations of credible, trustworthy members of your network. http://www.linkedin.com/e/gqo4xa-gn9v8ca1-1k/svp/inv-25/ -- (c) 2011, LinkedIn Corporation -------------- next part -------------- An HTML attachment was scrubbed... URL: From feng.tan at oracle.com Wed May 4 06:18:21 2011 From: feng.tan at oracle.com (feng.tan at oracle.com) Date: Tue, 3 May 2011 23:18:21 -0700 (PDT) Subject: Auto Reply: Invitation to connect on LinkedIn Message-ID: <7a3c4cb0-6d5b-4567-8dd6-fda84648f5cd@default> ???????????????? Dear Sender, >From May 3 ~ May 6, I am in Vacation, in this period I can not check email frequently. For Lab/Sending notification/Import/Export issue, please contact Carson Zhu(mailto:carson.zhu at oracle.com) or my manager Solon Zhang(mailto:solon.zhang at oracle.com). If there is any urgent case, you can reach me at 86-135-010-88029 Thanks, Tan Feng From beyonddc.storage at gmail.com Wed May 4 14:39:23 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Wed, 4 May 2011 10:39:23 -0400 Subject: Monitoring kickstart install status In-Reply-To: <2077006858-1304467149-cardhu_decombobulator_blackberry.rim.net-856587006-@bda614.bisx.prod.on.blackberry> References: <2077006858-1304467149-cardhu_decombobulator_blackberry.rim.net-856587006-@bda614.bisx.prod.on.blackberry> Message-ID: Hi DTK, Thanks for your feedback. Since we will be administrating all nodes deployment therefore my team concluded that we will not support group-installs. All packages will be explicitly listed in the %packages section. - David On Tue, May 3, 2011 at 7:59 PM, wrote: > Remember, in the ks, you can specify group-installs, and dependancies > should be resolved, so the number of packages in the ks will typically tend > to be low. Instead, look for specific marker packages to be installed, such > as kernel, or your standard userspace, or service-specific packages (apache, > if it is a web-server; dovecot, if it is a mail/groupware server; local > application package, etc) and send a syslog message/snmp trap/MQ message > when you see one of these markers. > > > -DTK > > > Sent via BlackBerry from T-Mobile > > -----Original Message----- > From: Chun Tat David Chu > Sender: kickstart-list-bounces at redhat.com > Date: Tue, 3 May 2011 17:34:43 > To: > Reply-To: Discussion list about Kickstart > Subject: Monitoring kickstart install status > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From beyonddc.storage at gmail.com Wed May 4 14:42:14 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Wed, 4 May 2011 10:42:14 -0400 Subject: Monitoring kickstart install status In-Reply-To: <4DC0976E.9070800@divms.uiowa.edu> References: <4DC0976E.9070800@divms.uiowa.edu> Message-ID: Hi Hugh, We have many machines that we will be building, so we are hoping to automate the process of monitoring the status of the OS. Spawning a script and monitoring 'install.log' is an option. Another thought would be monitoring the anaconda.log. I notice the log prints the following.... anaconda.log:21:22:31 INFO : moving (1) to step partitionobjinit anaconda.log:21:22:32 INFO : moving (1) to step autopartitionexecute anaconda.log:21:22:32 INFO : moving (1) to step partitiondone anaconda.log:21:22:32 INFO : moving (1) to step bootloadersetup anaconda.log:21:22:32 INFO : moving (1) to step networkdevicecheck anaconda.log:21:22:32 INFO : moving (1) to step reposetup anaconda.log:21:22:34 INFO : moving (1) to step basepkgsel anaconda.log:21:22:35 INFO : moving (1) to step postselection anaconda.log:21:22:44 INFO : moving (1) to step install anaconda.log:21:22:44 INFO : moving (1) to step enablefilesystems anaconda.log:21:23:48 INFO : moving (1) to step migratefilesystems anaconda.log:21:23:48 INFO : moving (1) to step setuptime anaconda.log:21:23:49 INFO : moving (1) to step preinstallconfig anaconda.log:21:23:49 INFO : moving (1) to step installpackages anaconda.log:21:25:57 INFO : moving (1) to step postinstallconfig anaconda.log:21:25:57 INFO : moving (1) to step writeconfig anaconda.log:21:25:57 INFO : moving (1) to step firstboot anaconda.log:21:25:57 INFO : moving (1) to step instbootloader anaconda.log:21:25:59 INFO : moving (1) to step writeksconfig anaconda.log:21:25:59 INFO : moving (1) to step writeregkey anaconda.log:21:25:59 INFO : moving (1) to step setfilecon anaconda.log:21:25:59 INFO : moving (1) to step copylogs It looks like if I monitor these steps then I might be able to get a better estimate of the installation progress. What do you think? Any thought? Thanks, David On Tue, May 3, 2011 at 8:01 PM, Hugh Brown wrote: > > > On 05/03/2011 04:34 PM, Chun Tat David Chu wrote: > >> Hi All, >> >> I am hoping to run a python script in the background during %pre to >> monitor >> the kickstart install status. My script will periodically send status to >> a >> server somewhere within LAN and report the current installation status. >> >> Running a python script in %pre and sending status to a remote server is >> the >> easy part but I am not sure what to monitor to retrieve the kickstart >> install status. >> >> One thought I have is to monitor the /mnt/sysimage/root/install.log and >> compare the number of packages installed and the number of packages >> specified in the ks.cfg and get a rough estimate of the kickstart install >> status. >> >> I am not sure if this is a correct way to do or there's a better to do so. >> >> Please let me know if you've any suggestion or thought. >> >> Thanks! >> >> - David >> > > My memory says that the install.log is actually in /tmp/install.log. After > the installation is finished it is copied to /mnt/sysimage/root/install.log. > > For fedora systems, I enabled sshd and require a password: > > sshpw --username=root XXXXXXXX --plaintext > > If I want to check status, I ssh in and look. If I don't, there's a script > that will email me when it has finished. > > On RH5 systems, I enabled vnc and would use vncviewer to connect and check > progress. > > Otherwise, watching the number of packages installed in install.log works. > > Hugh > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From hbrown at divms.uiowa.edu Wed May 4 16:53:55 2011 From: hbrown at divms.uiowa.edu (Hugh Brown) Date: Wed, 04 May 2011 11:53:55 -0500 Subject: Monitoring kickstart install status In-Reply-To: References: <4DC0976E.9070800@divms.uiowa.edu> Message-ID: <4DC184A3.1080503@divms.uiowa.edu> On 05/04/2011 09:42 AM, Chun Tat David Chu wrote: > Hi Hugh, > > We have many machines that we will be building, so we are hoping to automate > the process of monitoring the status of the OS. > > Spawning a script and monitoring 'install.log' is an option. Another > thought would be monitoring the anaconda.log. > > I notice the log prints the following.... > anaconda.log:21:22:31 INFO : moving (1) to step partitionobjinit > anaconda.log:21:22:32 INFO : moving (1) to step autopartitionexecute > anaconda.log:21:22:32 INFO : moving (1) to step partitiondone > anaconda.log:21:22:32 INFO : moving (1) to step bootloadersetup > anaconda.log:21:22:32 INFO : moving (1) to step networkdevicecheck > anaconda.log:21:22:32 INFO : moving (1) to step reposetup > anaconda.log:21:22:34 INFO : moving (1) to step basepkgsel > anaconda.log:21:22:35 INFO : moving (1) to step postselection > anaconda.log:21:22:44 INFO : moving (1) to step install > anaconda.log:21:22:44 INFO : moving (1) to step enablefilesystems > anaconda.log:21:23:48 INFO : moving (1) to step migratefilesystems > anaconda.log:21:23:48 INFO : moving (1) to step setuptime > anaconda.log:21:23:49 INFO : moving (1) to step preinstallconfig > anaconda.log:21:23:49 INFO : moving (1) to step installpackages > anaconda.log:21:25:57 INFO : moving (1) to step postinstallconfig > anaconda.log:21:25:57 INFO : moving (1) to step writeconfig > anaconda.log:21:25:57 INFO : moving (1) to step firstboot > anaconda.log:21:25:57 INFO : moving (1) to step instbootloader > anaconda.log:21:25:59 INFO : moving (1) to step writeksconfig > anaconda.log:21:25:59 INFO : moving (1) to step writeregkey > anaconda.log:21:25:59 INFO : moving (1) to step setfilecon > anaconda.log:21:25:59 INFO : moving (1) to step copylogs > > It looks like if I monitor these steps then I might be able to get a better > estimate of the installation progress. > > What do you think? Any thought? > > Thanks, > > David > How fine grained do you want the reporting to be? How long are your loads taking? From the steps above, you are doing a quick load and the system takes about 3-4 minutes before it's done. If all of your systems are going to be completed that quickly, then something in %post that does an "I'm done" would be sufficient. Our systems take about 1-2 hours to load (lots of packages and then a number of customizations). The bulk of our load time is spent in the package loading stage. So, you could poll anaconda.log for when it starts to install packages. Then switch to the install.log to track how far along in the package installation you are. Then switch back to anaconda.log for the final bits. Ultimately, we don't need a progress meter for the loads. So the "I'm done" is sufficient for us. If it seems like it is taking too long or I need to watch a load carefully, I ssh in and check on it. Hugh From beyonddc.storage at gmail.com Wed May 4 18:11:10 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Wed, 4 May 2011 14:11:10 -0400 Subject: Monitoring kickstart install status In-Reply-To: <4DC184A3.1080503@divms.uiowa.edu> References: <4DC0976E.9070800@divms.uiowa.edu> <4DC184A3.1080503@divms.uiowa.edu> Message-ID: Hi Hugh, Thanks for your feedback. Our reporting is hoping to be a percentage and the load time would probably be around 10 mins or so. I like your idea about monitoring both anaconda.log and install.log. Would putting something like "I'm done" in the last line of %post would ensure that's the last thing that is being executed by anaconda? or that's just close enough to being the last thing? Thanks for your valuable opinions. - David On Wed, May 4, 2011 at 12:53 PM, Hugh Brown wrote: > On 05/04/2011 09:42 AM, Chun Tat David Chu wrote: > >> Hi Hugh, >> >> We have many machines that we will be building, so we are hoping to >> automate >> the process of monitoring the status of the OS. >> >> Spawning a script and monitoring 'install.log' is an option. Another >> thought would be monitoring the anaconda.log. >> >> I notice the log prints the following.... >> anaconda.log:21:22:31 INFO : moving (1) to step partitionobjinit >> anaconda.log:21:22:32 INFO : moving (1) to step autopartitionexecute >> anaconda.log:21:22:32 INFO : moving (1) to step partitiondone >> anaconda.log:21:22:32 INFO : moving (1) to step bootloadersetup >> anaconda.log:21:22:32 INFO : moving (1) to step networkdevicecheck >> anaconda.log:21:22:32 INFO : moving (1) to step reposetup >> anaconda.log:21:22:34 INFO : moving (1) to step basepkgsel >> anaconda.log:21:22:35 INFO : moving (1) to step postselection >> anaconda.log:21:22:44 INFO : moving (1) to step install >> anaconda.log:21:22:44 INFO : moving (1) to step enablefilesystems >> anaconda.log:21:23:48 INFO : moving (1) to step migratefilesystems >> anaconda.log:21:23:48 INFO : moving (1) to step setuptime >> anaconda.log:21:23:49 INFO : moving (1) to step preinstallconfig >> anaconda.log:21:23:49 INFO : moving (1) to step installpackages >> anaconda.log:21:25:57 INFO : moving (1) to step postinstallconfig >> anaconda.log:21:25:57 INFO : moving (1) to step writeconfig >> anaconda.log:21:25:57 INFO : moving (1) to step firstboot >> anaconda.log:21:25:57 INFO : moving (1) to step instbootloader >> anaconda.log:21:25:59 INFO : moving (1) to step writeksconfig >> anaconda.log:21:25:59 INFO : moving (1) to step writeregkey >> anaconda.log:21:25:59 INFO : moving (1) to step setfilecon >> anaconda.log:21:25:59 INFO : moving (1) to step copylogs >> >> It looks like if I monitor these steps then I might be able to get a >> better >> estimate of the installation progress. >> >> What do you think? Any thought? >> >> Thanks, >> >> David >> >> > > How fine grained do you want the reporting to be? How long are your loads > taking? From the steps above, you are doing a quick load and the system > takes about 3-4 minutes before it's done. If all of your systems are going > to be completed that quickly, then something in %post that does an "I'm > done" would be sufficient. > > Our systems take about 1-2 hours to load (lots of packages and then a > number of customizations). The bulk of our load time is spent in the > package loading stage. So, you could poll anaconda.log for when it starts > to install packages. Then switch to the install.log to track how far along > in the package installation you are. Then switch back to anaconda.log for > the final bits. > > Ultimately, we don't need a progress meter for the loads. So the "I'm > done" is sufficient for us. If it seems like it is taking too long or I > need to watch a load carefully, I ssh in and check on it. > > > Hugh > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From hbrown at divms.uiowa.edu Wed May 4 18:23:31 2011 From: hbrown at divms.uiowa.edu (Hugh Brown) Date: Wed, 04 May 2011 13:23:31 -0500 Subject: Monitoring kickstart install status In-Reply-To: References: <4DC0976E.9070800@divms.uiowa.edu> <4DC184A3.1080503@divms.uiowa.edu> Message-ID: <4DC199A3.2090600@divms.uiowa.edu> On 05/04/2011 01:11 PM, Chun Tat David Chu wrote: > Hi Hugh, > > Thanks for your feedback. Our reporting is hoping to be a percentage and > the load time would probably be around 10 mins or so. I like your idea > about monitoring both anaconda.log and install.log. > > Would putting something like "I'm done" in the last line of %post would > ensure that's the last thing that is being executed by anaconda? or that's > just close enough to being the last thing? > > Thanks for your valuable opinions. > > - David > The scripts that are run from %post are effectively the last thing that happens (that we have any control over). So you could put another python script or shell script in %post that would contact your monitoring server to indicate completion. For example, in looking at a machine that just loaded, the last line in anaconda.log is time stamped 12:07:18. The log files that I create from scripts within %post are timestamped from 12:07 to 12:14. The next available time to indicate completion would be to drop a "firstboot" style script into the newly installed machine and have it report after the reboot. After it reports, it could remove/disable itself. Hugh From beyonddc.storage at gmail.com Wed May 4 20:41:04 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Wed, 4 May 2011 16:41:04 -0400 Subject: Monitoring kickstart install status In-Reply-To: <4DC199A3.2090600@divms.uiowa.edu> References: <4DC0976E.9070800@divms.uiowa.edu> <4DC184A3.1080503@divms.uiowa.edu> <4DC199A3.2090600@divms.uiowa.edu> Message-ID: Hi Hugh, Thanks for your input. I will toy with your idea in my environment. Thanks again! - David On Wed, May 4, 2011 at 2:23 PM, Hugh Brown wrote: > On 05/04/2011 01:11 PM, Chun Tat David Chu wrote: > >> Hi Hugh, >> >> Thanks for your feedback. Our reporting is hoping to be a percentage and >> the load time would probably be around 10 mins or so. I like your idea >> about monitoring both anaconda.log and install.log. >> >> Would putting something like "I'm done" in the last line of %post would >> ensure that's the last thing that is being executed by anaconda? or that's >> just close enough to being the last thing? >> >> Thanks for your valuable opinions. >> >> - David >> >> > > The scripts that are run from %post are effectively the last thing that > happens (that we have any control over). So you could put another python > script or shell script in %post that would contact your monitoring server to > indicate completion. > > For example, in looking at a machine that just loaded, the last line in > anaconda.log is time stamped 12:07:18. The log files that I create from > scripts within %post are timestamped from 12:07 to 12:14. > > The next available time to indicate completion would be to drop a > "firstboot" style script into the newly installed machine and have it report > after the reboot. After it reports, it could remove/disable itself. > > > Hugh > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jason at rampaginggeek.com Wed May 4 23:59:00 2011 From: jason at rampaginggeek.com (Jason Edgecombe) Date: Wed, 04 May 2011 19:59:00 -0400 Subject: Monitoring kickstart install status In-Reply-To: References: <4DC0976E.9070800@divms.uiowa.edu> <4DC184A3.1080503@divms.uiowa.edu> <4DC199A3.2090600@divms.uiowa.edu> Message-ID: <4DC1E844.4010506@rampaginggeek.com> kickstart can log to a remote syslog daemon. Would that be OK? You could configure /etc/syslog.conf in %post so that the system logs after first boot as well. To log extra stuff, just use the "logger" command. I'm not sure if "logger" is available during installation, though, but you could copy it in if needed. Jason On 05/04/2011 04:41 PM, Chun Tat David Chu wrote: > Hi Hugh, > > Thanks for your input. I will toy with your idea in my environment. > > Thanks again! > > - David > > On Wed, May 4, 2011 at 2:23 PM, Hugh Brown wrote: > >> On 05/04/2011 01:11 PM, Chun Tat David Chu wrote: >> >>> Hi Hugh, >>> >>> Thanks for your feedback. Our reporting is hoping to be a percentage and >>> the load time would probably be around 10 mins or so. I like your idea >>> about monitoring both anaconda.log and install.log. >>> >>> Would putting something like "I'm done" in the last line of %post would >>> ensure that's the last thing that is being executed by anaconda? or that's >>> just close enough to being the last thing? >>> >>> Thanks for your valuable opinions. >>> >>> - David >>> >>> >> The scripts that are run from %post are effectively the last thing that >> happens (that we have any control over). So you could put another python >> script or shell script in %post that would contact your monitoring server to >> indicate completion. >> >> For example, in looking at a machine that just loaded, the last line in >> anaconda.log is time stamped 12:07:18. The log files that I create from >> scripts within %post are timestamped from 12:07 to 12:14. >> >> The next available time to indicate completion would be to drop a >> "firstboot" style script into the newly installed machine and have it report >> after the reboot. After it reports, it could remove/disable itself. >> >> >> Hugh >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list >> > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From Floydsmith at aol.com Thu May 5 06:29:48 2011 From: Floydsmith at aol.com (Floydsmith at aol.com) Date: Thu, 5 May 2011 02:29:48 -0400 (EDT) Subject: beta fc 15 strange read error on USB stick drive Message-ID: <19d74.4a80cc70.3af39ddb@aol.com> I downloaded the Beta on May 1. It uses anaconda 15.27 (11:555:41). I boot off a sandisk FAT32 32GB USB stick (which has a single partition) and with my kickstart file is on that drive. Intermittantly (about 50% of the time over the last 6 times I have tried) immediately after the probing the ethernet Network device screen (device is from my kickstart file) (and hence AFTER my kickstart file has been read in) I get a screen that says: My sandisk drive may contain data... We could not detect partitions... And it presents a box which says: Apply my choice to all such devices (Which I check) And below that two choices: Yes, discard data... No, keep data... Which I select. The Examining Storage Devices screen then appears and the install always proceeds and finishes OK after that. Is there any way I can specify these two selections in my kickstart file. I don't ZERO my MSB and I don't clear the partions (just use a single onpart to format an existing ext3 partition) so it seems that it should be able to figure that I don't want any other data destroyed (and not the one my kickstart file is on). Any help will be greatly appreciated. Floyd, -------------- next part -------------- An HTML attachment was scrubbed... URL: From srobson at cadence.com Thu May 5 16:47:42 2011 From: srobson at cadence.com (Steve Robson) Date: Thu, 5 May 2011 17:47:42 +0100 Subject: beta fc 15 strange read error on USB stick drive In-Reply-To: References: Message-ID: <4DC2D4AE.6080909@cadence.com> floydsmith at aol.com wrote: > > I downloaded the Beta on May 1. It uses anaconda 15.27 (11:555:41). I > boot off > a sandisk FAT32 32GB USB stick (which has a single partition) and with my > kickstart file is on that drive. > Intermittantly (about 50% of the time over the last 6 times I have tried) > immediately after the probing the ethernet Network device screen (device > is from > my kickstart file) (and hence AFTER my kickstart file has been read in) > I get > a screen that says: > My sandisk drive may contain data... We could not detect partitions... > And it presents a box which says: > Apply my choice to all such devices > (Which I check) > And below that two choices: > Yes, discard data... > No, keep data... > Which I select. > The Examining Storage Devices screen then appears and the install always > proceeds and finishes OK after that. > Is there any way I can specify these two selections in my kickstart file. > I don't ZERO my MSB and I don't clear the partions (just use a single onpart > to format an existing ext3 partition) so it seems that it should be able to > figure that I don't want any other data destroyed (and not the one my > kickstart file is on). > Any help will be greatly appreciated. > Floyd, Not sure if it's FC-compatible but the "ignoredisk" kickstart directive might be suitable. Something like: ignoredisk --drives=drive1,drive2,... where driveN is one of sda, sdb,..., hda,... etc. or: ignoredisk --only-use=sda which specifies a list of disks for the installer to use. All other disks are ignored. For example, to use disk sda during installation and ignore all other disks. Lifted from http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Installation_Guide/s1-kickstart2-options.html -- Regards, Steve IT Support - UNIX/Linux Cadence Design Systems Bagshot Road Bracknell BERKSHIRE RG12 0PH UK From beyonddc.storage at gmail.com Thu May 5 19:52:06 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Thu, 5 May 2011 15:52:06 -0400 Subject: Question about output in the tmp/install.log Message-ID: Hi All, I have been playing with anaconda for the last couple of days and there's one thing that I couldn't explain. In my kickstart file, I specified a package to install using the fully qualified name... for example "mktemp-1.5-23.2.2.x86_64" When I review the install.log after the system is build, the package that was installing is named "mktemp-3:1.5-23.2.2.x86_64" instead of "mktemp-1.5-23.2.2.x86_64". When I use rpm to query for that particular package, the package version that get installed is indeed "mktemp-1.5-23.2.2.x86_64" Do you know why there's an extra "3:" in the package name in the install.log? Thanks! - David -------------- next part -------------- An HTML attachment was scrubbed... URL: From clumens at redhat.com Thu May 5 20:02:56 2011 From: clumens at redhat.com (Chris Lumens) Date: Thu, 5 May 2011 16:02:56 -0400 Subject: Question about output in the tmp/install.log In-Reply-To: References: Message-ID: <20110505200255.GN18330@exeter.usersys.redhat.com> > Do you know why there's an extra "3:" in the package name in the > install.log? That's the package's epoch (sort of like a super-version number). Not all packages have one. Not all utilities choose to print it out even if the package has one. - Chris From beyonddc.storage at gmail.com Thu May 5 20:18:23 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Thu, 5 May 2011 16:18:23 -0400 Subject: Question about output in the tmp/install.log In-Reply-To: <20110505200255.GN18330@exeter.usersys.redhat.com> References: <20110505200255.GN18330@exeter.usersys.redhat.com> Message-ID: Got it, thanks! On Thu, May 5, 2011 at 4:02 PM, Chris Lumens wrote: > > Do you know why there's an extra "3:" in the package name in the > > install.log? > > That's the package's epoch (sort of like a super-version number). Not > all packages have one. Not all utilities choose to print it out even if > the package has one. > > - Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From andy.speagle at wichita.edu Mon May 9 14:14:21 2011 From: andy.speagle at wichita.edu (Speagle, Andy) Date: Mon, 9 May 2011 09:14:21 -0500 Subject: Use entire disk as PV. In-Reply-To: References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> Message-ID: Hi Team, I?ve progressed a bit on this issue. What I?m presently trying to do is use a simple %pre script to generate a temp file with the commands to be included in the partitioning section of the kickstart. So, I run the %pre script, it creates /tmp/custom-part, which is included in the partitioning section using: %include /tmp/custom-part All of that works duckily? however anaconda refuses to use ?rootvg? without it being defined using the ?volgroup? command in the partitioning section. And sadly, I can?t get the ?volgroup? command to work without a PV defined. If I try to specify the partition by passing the --useexisting and --onpart options to the ?part? command? that fails too ? so I continue to be stuck. This is what I do to have the /tmp/custom-part created: echo "part /boot --fstype=ext3 --onpart=$BOOT" > /tmp/custom-part echo "volgroup rootvg --useexisting" >> /tmp/custom-part echo "logvol / --fstype=ext3 --name=root --vgname=rootvg --size=4096" >> /tmp/custom-part echo "logvol swap --fstype=swap --name=swap --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /var --fstype=ext3 --name=var --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /var/tmp --fstype=ext3 --name=vartmp --vgname=rootvg --size=1024" >> /tmp/custom-part echo "logvol /home --fstype=ext3 --name=home --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /tmp --fstype=ext3 --name=tmp --vgname=rootvg --size=1024" >> /tmp/custom-part echo "logvol /opt --fstype=ext3 --name=opt --vgname=rootvg --size=2048" >> /tmp/custom-part Has anyone actually done this? Is there some magic I?m missing? -Andy From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Callahan, Tom Sent: Tuesday, May 03, 2011 5:45 AM To: Discussion list about Kickstart Cc: Discussion list about Kickstart Subject: Re: Use entire disk as PV. Most storage arrays can grow existing disks, which doesn't work as nicely if the disk has a partition. Making a PV out of the entire disk allows much easier PV/VG/LV expansion, without mucking with a partition table. As for the training piece, I'd expect an admin would do some verification a disk is not in use before blindly assuming it's not in use, and overwriting data. There is no standard practice with how to create a PV, it depends on the site. On May 3, 2011, at 5:30 AM, "Moray Henderson" > wrote: From: Speagle, Andy [mailto:andy.speagle at wichita.edu] I?m trying to setup the partitioning section of my kickstarts in such a way that rather than partitioning a disk and using /dev/sdX1 as the PV for my root VG, that I can instead use the entire disk. The reason for doing this would be to make PV resizing a bit easier for virtual machines. Otherwise, I must muck around with the partition table and blah blah ? not a show-stopper, but potentially dangerous for junior admins. I suspect that the functionality just isn?t there? but does anyone know the magic syntax to get it to use the entire drive (aka. /dev/sdb) as a PV during kickstart? Why do you need to resize PVs for virtual machines? You allocate the extra space to your VM as another virtual drive or partition, then using LVM create another Physical Volume for that drive, add that Physical Volume to your Volume Group and Logical Volume, and extend the filesystem. No mucking around with an existing partition table. I would be confused if I encountered a disk without a partition table ? I?d probably assume it wasn?t formatted at all. Better to train your junior admins in standard practice? Moray. ?To err is human; to purr, feline.? _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From Moray.Henderson at om.org Mon May 9 16:39:20 2011 From: Moray.Henderson at om.org (Moray Henderson (ICT)) Date: Mon, 9 May 2011 17:39:20 +0100 Subject: Use entire disk as PV. In-Reply-To: References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> Message-ID: <001c01cc0e67$a6bac4a0$f4304de0$@Henderson@om.org> As well as generating /tmp/custom-part for kickstart, your %pre script also has to execute the Linux commands to create the PV. There were some examples given earlier in this thread: lvm pvcreate /dev/sdb echo "volgroup rootvg /dev/sdb ?useexisting" >> /tmp/custom-part ought to give you a volgroup called rootvg consisting of the whole of /dev/sdb. See 31.4. Kickstart Options if you haven?t found it already. Moray. ?To err is human; to purr, feline.? From: Speagle, Andy [mailto:andy.speagle at wichita.edu] Sent: 09 May 2011 15:14 To: Discussion list about Kickstart Subject: RE: Use entire disk as PV. Hi Team, I?ve progressed a bit on this issue. What I?m presently trying to do is use a simple %pre script to generate a temp file with the commands to be included in the partitioning section of the kickstart. So, I run the %pre script, it creates /tmp/custom-part, which is included in the partitioning section using: %include /tmp/custom-part All of that works duckily? however anaconda refuses to use ?rootvg? without it being defined using the ?volgroup? command in the partitioning section. And sadly, I can?t get the ?volgroup? command to work without a PV defined. If I try to specify the partition by passing the --useexisting and --onpart options to the ?part? command? that fails too ? so I continue to be stuck. This is what I do to have the /tmp/custom-part created: echo "part /boot --fstype=ext3 --onpart=$BOOT" > /tmp/custom-part echo "volgroup rootvg --useexisting" >> /tmp/custom-part echo "logvol / --fstype=ext3 --name=root --vgname=rootvg --size=4096" >> /tmp/custom-part echo "logvol swap --fstype=swap --name=swap --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /var --fstype=ext3 --name=var --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /var/tmp --fstype=ext3 --name=vartmp --vgname=rootvg --size=1024" >> /tmp/custom-part echo "logvol /home --fstype=ext3 --name=home --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /tmp --fstype=ext3 --name=tmp --vgname=rootvg --size=1024" >> /tmp/custom-part echo "logvol /opt --fstype=ext3 --name=opt --vgname=rootvg --size=2048" >> /tmp/custom-part Has anyone actually done this? Is there some magic I?m missing? -Andy ________________________________ OM International Limited - Unit B Clifford Court, Cooper Way - Carlisle CA3 0JG - United Kingdom Charity reg no: 1112655 - Company reg no: 5649412 (England and Wales) -------------- next part -------------- An HTML attachment was scrubbed... URL: From andy.speagle at wichita.edu Mon May 9 18:33:50 2011 From: andy.speagle at wichita.edu (Speagle, Andy) Date: Mon, 9 May 2011 13:33:50 -0500 Subject: Use entire disk as PV. In-Reply-To: <001c01cc0e67$a6bac4a0$f4304de0$%Henderson@om.org> References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> <001c01cc0e67$a6bac4a0$f4304de0$%Henderson@om.org> Message-ID: I?ve actually tried that in the past. It doesn?t work. But, for completeness, I tried it again? and I receive this error from anaconda: Error Parsing Kickstart Config The following error was found while parsing the kickstart configuration file: The following problem occurred on line 2 of the kickstart file: Tried to use undefined partition /dev/sdb in Volume Group specification Any thoughts on that? I?m at a loss. -Andy From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Moray Henderson (ICT) Sent: Monday, May 09, 2011 11:39 AM To: 'Discussion list about Kickstart' Subject: RE: Use entire disk as PV. As well as generating /tmp/custom-part for kickstart, your %pre script also has to execute the Linux commands to create the PV. There were some examples given earlier in this thread: lvm pvcreate /dev/sdb echo "volgroup rootvg /dev/sdb ?useexisting" >> /tmp/custom-part ought to give you a volgroup called rootvg consisting of the whole of /dev/sdb. See 31.4. Kickstart Options if you haven?t found it already. Moray. ?To err is human; to purr, feline.? From: Speagle, Andy [mailto:andy.speagle at wichita.edu] Sent: 09 May 2011 15:14 To: Discussion list about Kickstart Subject: RE: Use entire disk as PV. Hi Team, I?ve progressed a bit on this issue. What I?m presently trying to do is use a simple %pre script to generate a temp file with the commands to be included in the partitioning section of the kickstart. So, I run the %pre script, it creates /tmp/custom-part, which is included in the partitioning section using: %include /tmp/custom-part All of that works duckily? however anaconda refuses to use ?rootvg? without it being defined using the ?volgroup? command in the partitioning section. And sadly, I can?t get the ?volgroup? command to work without a PV defined. If I try to specify the partition by passing the --useexisting and --onpart options to the ?part? command? that fails too ? so I continue to be stuck. This is what I do to have the /tmp/custom-part created: echo "part /boot --fstype=ext3 --onpart=$BOOT" > /tmp/custom-part echo "volgroup rootvg --useexisting" >> /tmp/custom-part echo "logvol / --fstype=ext3 --name=root --vgname=rootvg --size=4096" >> /tmp/custom-part echo "logvol swap --fstype=swap --name=swap --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /var --fstype=ext3 --name=var --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /var/tmp --fstype=ext3 --name=vartmp --vgname=rootvg --size=1024" >> /tmp/custom-part echo "logvol /home --fstype=ext3 --name=home --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /tmp --fstype=ext3 --name=tmp --vgname=rootvg --size=1024" >> /tmp/custom-part echo "logvol /opt --fstype=ext3 --name=opt --vgname=rootvg --size=2048" >> /tmp/custom-part Has anyone actually done this? Is there some magic I?m missing? -Andy ________________________________ OM International Limited - Unit B Clifford Court, Cooper Way - Carlisle CA3 0JG - United Kingdom Charity reg no: 1112655 - Company reg no: 5649412 (England and Wales) -------------- next part -------------- An HTML attachment was scrubbed... URL: From Moray.Henderson at ict-software.org Tue May 10 09:02:52 2011 From: Moray.Henderson at ict-software.org (Moray Henderson) Date: Tue, 10 May 2011 10:02:52 +0100 Subject: Use entire disk as PV. In-Reply-To: References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> <001c01cc0e67$a6bac4a0$f4304de0$%Henderson@om.org> Message-ID: <000001cc0ef1$0c82fd00$2588f700$@Henderson@ict-software.org> Did you use a double dash for ?--preexisting?? I didn?t notice as I was typing it that Outlook had ?helpfully? changed the two dashes to a single emdash. Did the pvcreate command actually work: does that disk now have a full-disk PV on it? In the pvcreate documentation, I just came across For whole disk devices only the partition table must be erased, which will effectively destroy all data on that disk. This can be done by zeroing the first sector with: dd if=/dev/zero of=PhysicalVolume bs=512 count=1 Here?s something else to try: echo "part pv.01 --onpart sdb --grow" >> /tmp/custom-part echo "volgroup rootvg /dev/sdb pv.01" >> /tmp/custom-part If kickstart refuses to create a partition on sdb, you will have to do something like lvm pvcreate /dev/sdb lvm vgcreate rootvg /dev/sdb echo "logvol / --fstype ext3 --name=root --vgname=rootvg --size=4096 --useexisting" >> /tmp/custom-part Moray. ?To err is human; to purr, feline.? From: Speagle, Andy [mailto:andy.speagle at wichita.edu] Sent: 09 May 2011 19:34 To: Discussion list about Kickstart Subject: RE: Use entire disk as PV. I?ve actually tried that in the past. It doesn?t work. But, for completeness, I tried it again? and I receive this error from anaconda: Error Parsing Kickstart Config The following error was found while parsing the kickstart configuration file: The following problem occurred on line 2 of the kickstart file: Tried to use undefined partition /dev/sdb in Volume Group specification Any thoughts on that? I?m at a loss. -Andy From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Moray Henderson (ICT) Sent: Monday, May 09, 2011 11:39 AM To: 'Discussion list about Kickstart' Subject: RE: Use entire disk as PV. As well as generating /tmp/custom-part for kickstart, your %pre script also has to execute the Linux commands to create the PV. There were some examples given earlier in this thread: lvm pvcreate /dev/sdb echo "volgroup rootvg /dev/sdb ?useexisting" >> /tmp/custom-part ought to give you a volgroup called rootvg consisting of the whole of /dev/sdb. See 31.4. Kickstart Options if you haven?t found it already. Moray. ?To err is human; to purr, feline.? From: Speagle, Andy [mailto:andy.speagle at wichita.edu] Sent: 09 May 2011 15:14 To: Discussion list about Kickstart Subject: RE: Use entire disk as PV. Hi Team, I?ve progressed a bit on this issue. What I?m presently trying to do is use a simple %pre script to generate a temp file with the commands to be included in the partitioning section of the kickstart. So, I run the %pre script, it creates /tmp/custom-part, which is included in the partitioning section using: %include /tmp/custom-part All of that works duckily? however anaconda refuses to use ?rootvg? without it being defined using the ?volgroup? command in the partitioning section. And sadly, I can?t get the ?volgroup? command to work without a PV defined. If I try to specify the partition by passing the --useexisting and --onpart options to the ?part? command? that fails too ? so I continue to be stuck. This is what I do to have the /tmp/custom-part created: echo "part /boot --fstype=ext3 --onpart=$BOOT" > /tmp/custom-part echo "volgroup rootvg --useexisting" >> /tmp/custom-part echo "logvol / --fstype=ext3 --name=root --vgname=rootvg --size=4096" >> /tmp/custom-part echo "logvol swap --fstype=swap --name=swap --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /var --fstype=ext3 --name=var --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /var/tmp --fstype=ext3 --name=vartmp --vgname=rootvg --size=1024" >> /tmp/custom-part echo "logvol /home --fstype=ext3 --name=home --vgname=rootvg --size=2048" >> /tmp/custom-part echo "logvol /tmp --fstype=ext3 --name=tmp --vgname=rootvg --size=1024" >> /tmp/custom-part echo "logvol /opt --fstype=ext3 --name=opt --vgname=rootvg --size=2048" >> /tmp/custom-part Has anyone actually done this? Is there some magic I?m missing? -Andy _____ OM International Limited - Unit B Clifford Court, Cooper Way - Carlisle CA3 0JG - United Kingdom Charity reg no: 1112655 - Company reg no: 5649412 (England and Wales) -------------- next part -------------- An HTML attachment was scrubbed... URL: From andy.speagle at wichita.edu Tue May 10 12:52:41 2011 From: andy.speagle at wichita.edu (Speagle, Andy) Date: Tue, 10 May 2011 07:52:41 -0500 Subject: Use entire disk as PV. In-Reply-To: <000001cc0ef1$0c82fd00$2588f700$%Henderson@ict-software.org> References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> <001c01cc0e67$a6bac4a0$f4304de0$%Henderson@om.org> <000001cc0ef1$0c82fd00$2588f700$%Henderson@ict-software.org> Message-ID: > Did you use a double dash for ?--preexisting??? I didn?t notice as I was typing it that Outlook had > ?helpfully? changed the two dashes to a single emdash.? Did the pvcreate command actually work: does that > disk now have a full-disk PV on it?? In the pvcreate documentation, I just came across Aye, I had used the --useexisting option with the double-hyphen. > Here?s something else to try: > > echo "part pv.01 --onpart sdb --grow" >> /tmp/custom-part > echo "volgroup rootvg /dev/sdb pv.01" >> /tmp/custom-part > > If kickstart refuses to create a partition on sdb, you will have to do something like > > lvm pvcreate /dev/sdb > lvm vgcreate rootvg /dev/sdb > echo "logvol / --fstype ext3 --name=root --vgname=rootvg --size=4096 --useexisting" >> /tmp/custom-part This seems to have worked for the RHEL 6 installer. RHEL 5.5 still barfs on this saying that no VG exists with the name "rootvg" ... I have filed a BZ with this as a feature request. I hope this gets added at some point. It should be fairly trivial to do. Thanks for the info. -Andy From Moray.Henderson at ict-software.org Tue May 10 13:11:35 2011 From: Moray.Henderson at ict-software.org (Moray Henderson) Date: Tue, 10 May 2011 14:11:35 +0100 Subject: Use entire disk as PV. In-Reply-To: References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> <001c01cc0e67$a6bac4a0$f4304de0$%Henderson@om.org> <000001cc0ef1$0c82fd00$2588f700$%Henderson@ict-software.org> Message-ID: <001001cc0f13$cb269380$6173ba80$@Henderson@ict-software.org> Speagle, Andy wrote: > Sent: 10 May 2011 13:53 > > Did you use a double dash for ?--preexisting?? I didn?t notice as I > was typing it that Outlook had > > ?helpfully? changed the two dashes to a single emdash. Did the > pvcreate command actually work: does that > > disk now have a full-disk PV on it? In the pvcreate documentation, I > just came across > > Aye, I had used the --useexisting option with the double-hyphen. > > > Here?s something else to try: > > > > echo "part pv.01 --onpart sdb --grow" >> /tmp/custom-part > > echo "volgroup rootvg /dev/sdb pv.01" >> /tmp/custom-part > > > > If kickstart refuses to create a partition on sdb, you will have to > do something like > > > > lvm pvcreate /dev/sdb > > lvm vgcreate rootvg /dev/sdb > > echo "logvol / --fstype ext3 --name=root --vgname=rootvg --size=4096 > --useexisting" >> /tmp/custom-part > > This seems to have worked for the RHEL 6 installer. RHEL 5.5 still > barfs on this saying that no VG exists with the name "rootvg" ... > > I have filed a BZ with this as a feature request. I hope this gets > added at some point. It should be fairly trivial to do. Is it any better in RHEL 5.6? Were you able to tell if the lvm commands in the %pre script actually worked? Try them manually from a kickstart shell; that may give a clue where things are going wrong. If the volgroup is being created, maybe kickstart in RHEL 5.5 isn't re-reading the disk information. Maybe there's some way to get it to do that, or at worst get your %pre script to detect whether the disk has volgroup information. If not, create it and reboot; else write /tmp/custom-part to create the logical volumes and continue the install. Moray. ?To err is human; to purr, feline.? From andy.speagle at wichita.edu Tue May 10 17:36:34 2011 From: andy.speagle at wichita.edu (Speagle, Andy) Date: Tue, 10 May 2011 12:36:34 -0500 Subject: Use entire disk as PV. In-Reply-To: <001001cc0f13$cb269380$6173ba80$%Henderson@ict-software.org> References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> <001c01cc0e67$a6bac4a0$f4304de0$%Henderson@om.org> <000001cc0ef1$0c82fd00$2588f700$%Henderson@ict-software.org> <001001cc0f13$cb269380$6173ba80$%Henderson@ict-software.org> Message-ID: > Speagle, Andy wrote: > Sent: 10 May 2011 13:53 > > > Did you use a double dash for ?--preexisting?? I didn?t notice as I > > was typing it that Outlook had > > > ?helpfully? changed the two dashes to a single emdash. Did the > > pvcreate command actually work: does that > > > disk now have a full-disk PV on it? In the pvcreate documentation, > > > I > > just came across > > > > Aye, I had used the --useexisting option with the double-hyphen. > > > > > Here?s something else to try: > > > > > > echo "part pv.01 --onpart sdb --grow" >> /tmp/custom-part echo > > > "volgroup rootvg /dev/sdb pv.01" >> /tmp/custom-part > > > > > > If kickstart refuses to create a partition on sdb, you will have to > > do something like > > > > > > lvm pvcreate /dev/sdb > > > lvm vgcreate rootvg /dev/sdb > > > echo "logvol / --fstype ext3 --name=root --vgname=rootvg --size=4096 > > --useexisting" >> /tmp/custom-part > > > > This seems to have worked for the RHEL 6 installer. RHEL 5.5 still > > barfs on this saying that no VG exists with the name "rootvg" ... > > > > I have filed a BZ with this as a feature request. I hope this gets > > added at some point. It should be fairly trivial to do. > > Is it any better in RHEL 5.6? > Were you able to tell if the lvm commands in the %pre script actually > worked? Try them manually from a kickstart shell; that may give a clue > where things are going wrong. If the volgroup is being created, maybe > kickstart in RHEL 5.5 isn't re-reading the disk information. Maybe > there's some way to get it to do that, or at worst get your %pre script to > detect whether the disk has volgroup information. If not, create it and > reboot; else write /tmp/custom-part to create the logical volumes and > continue the install. RHEL 5.6 suffers the same fate. I received a response on my BZ for this feature request that native support for PVs on entire disk will be included for anaconda somewhere in RHEL 6 ... but that it will not be back-ported to any RHEL 5 anaconda. All of the LVM commands are working just duckily. Anaconda just refuses to handle this situation. The error is quite explicit that it's an issue parsing the kickstart file. Anaconda is expecting a "volgroup" definition and won't continue without one. -Andy From darn570-linux at yahoo.com Sun May 15 16:43:13 2011 From: darn570-linux at yahoo.com (Chad Darnielle) Date: Sun, 15 May 2011 09:43:13 -0700 (PDT) Subject: Use entire disk as PV. References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> <001c01cc0e67$a6bac4a0$f4304de0$%Henderson@om.org> <000001cc0ef1$0c82fd00$2588f700$%Henderson@ict-software.org> <001001cc0f13$cb269380$6173ba80$%Henderson@ict-software.org> Message-ID: <613014.23680.qm@web112914.mail.gq1.yahoo.com> Andy, I have similar requirements and have went in search of how to do this through kickstart/anaconda. For RHEL 5 I ended up using vmware templates which was something I had hoped to avoid so it's good to hear that hopefully with RHEL 6 I might be able to get away from this at least with %pre functionality. My Red Hat technical contact was supposed to file through internal channels an RFE on this as well which I need to check on. Do you mind sharing the BZ# you submitted so I can reference this as well. I haven't had a chance to see if the rescue environment has caught up with this practice or not.? The reason I had looked at the whole disk model was for our virtualized environment as well on VMware. We usually do disk/file system expansions?dynamically. With RHEL 5 on VMware?I attempted to get away from?our historical model of addind a new LUN, partition, and then add to LVM, etc. There was also some performance recommendations from VMware to keep the number of virtual disks low per VMFS volume and other VM sizing considerations (# of LUN's per virtual storage controller). For my VM's I am essentially creating a single 512m partitioned disk for /boot, and then a large unpartitioned disk for the LVM PV which hosted everything else. Whenever I needed to increase space I would resize the second LUN in VMware, poke the correct?sysfs items, &?then run a pvresize, etc. I'm assuming I can probably do the samething with partitions by resizing the partition (assuming it's at the end of the disk) that hosts the PV but I didn't have any extensive experience with?resizing partitions. I have had?past issues with changes to the partition table requiring reboots so I shyed?away from really exploring this too deeply based on the amount of time I had. Altering the partition table on a live system also gives me and other system admins here the creeps :-). Chad ________________________________ From: "Speagle, Andy" To: Discussion list about Kickstart Sent: Tuesday, May 10, 2011 12:36 PM Subject: RE: Use entire disk as PV. > Speagle, Andy wrote: > Sent: 10 May 2011 13:53 > > > Did you use a double dash for ?--preexisting??? I didn?t notice as I > > was typing it that Outlook had > > > ?helpfully? changed the two dashes to a single emdash.? Did the > > pvcreate command actually work: does that > > > disk now have a full-disk PV on it?? In the pvcreate documentation, > > > I > > just came across > > > > Aye, I had used the --useexisting option with the double-hyphen. > > > > > Here?s something else to try: > > > > > > echo "part pv.01 --onpart sdb --grow" >> /tmp/custom-part echo > > > "volgroup rootvg /dev/sdb pv.01" >> /tmp/custom-part > > > > > > If kickstart refuses to create a partition on sdb, you will have to > > do something like > > > > > > lvm pvcreate /dev/sdb > > > lvm vgcreate rootvg /dev/sdb > > > echo "logvol / --fstype ext3 --name=root --vgname=rootvg --size=4096 > > --useexisting" >> /tmp/custom-part > > > > This seems to have worked for the RHEL 6 installer.? RHEL 5.5 still > > barfs on this saying that no VG exists with the name "rootvg" ... > > > > I have filed a BZ with this as a feature request.? I hope this gets > > added at some point.? It should be fairly trivial to do. > > Is it any better in RHEL 5.6? > Were you able to tell if the lvm commands in the %pre script actually > worked?? Try them manually from a kickstart shell; that may give a clue > where things are going wrong.? If the volgroup is being created, maybe > kickstart in RHEL 5.5 isn't re-reading the disk information.? Maybe > there's some way to get it to do that, or at worst get your %pre script to > detect whether the disk has volgroup information.? If not, create it and > reboot; else write /tmp/custom-part to create the logical volumes and > continue the install. RHEL 5.6 suffers the same fate.? I received a response on my BZ for this feature request that native support for PVs on entire disk will be included for anaconda somewhere in RHEL 6 ... but that it will not be back-ported to any RHEL 5 anaconda.? All of the LVM commands are working just duckily.? Anaconda just refuses to handle this situation.? The error is quite explicit that it's an issue parsing the kickstart file.? Anaconda is expecting a "volgroup" definition and won't continue without one. -Andy _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From andy.speagle at wichita.edu Sun May 15 20:42:55 2011 From: andy.speagle at wichita.edu (Speagle, Andy) Date: Sun, 15 May 2011 15:42:55 -0500 Subject: Use entire disk as PV. In-Reply-To: <613014.23680.qm@web112914.mail.gq1.yahoo.com> References: <000501cc0971$e1033520$a3099f60$%Henderson@ict-software.org> <001c01cc0e67$a6bac4a0$f4304de0$%Henderson@om.org> <000001cc0ef1$0c82fd00$2588f700$%Henderson@ict-software.org> <001001cc0f13$cb269380$6173ba80$%Henderson@ict-software.org> <613014.23680.qm@web112914.mail.gq1.yahoo.com> Message-ID: Ok, Firstly? https://bugzilla.redhat.com/show_bug.cgi?id=703161 It looks like it?s been closed. The word is that the whole disk formats are recognized and used during kickstarts, but they?re not in a hurry to give a UI feature for creating them? so the %pre script will need to remain. I don?t mind mucking with the partition table in order to extend the PV ? but I?d prefer not to. Ah well. -Andy From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Chad Darnielle Sent: Sunday, May 15, 2011 11:43 AM To: Discussion list about Kickstart Subject: Re: Use entire disk as PV. Andy, I have similar requirements and have went in search of how to do this through kickstart/anaconda. For RHEL 5 I ended up using vmware templates which was something I had hoped to avoid so it's good to hear that hopefully with RHEL 6 I might be able to get away from this at least with %pre functionality. My Red Hat technical contact was supposed to file through internal channels an RFE on this as well which I need to check on. Do you mind sharing the BZ# you submitted so I can reference this as well. I haven't had a chance to see if the rescue environment has caught up with this practice or not. The reason I had looked at the whole disk model was for our virtualized environment as well on VMware. We usually do disk/file system expansions dynamically. With RHEL 5 on VMware I attempted to get away from our historical model of addind a new LUN, partition, and then add to LVM, etc. There was also some performance recommendations from VMware to keep the number of virtual disks low per VMFS volume and other VM sizing considerations (# of LUN's per virtual storage controller). For my VM's I am essentially creating a single 512m partitioned disk for /boot, and then a large unpartitioned disk for the LVM PV which hosted everything else. Whenever I needed to increase space I would resize the second LUN in VMware, poke the correct sysfs items, & then run a pvresize, etc. I'm assuming I can probably do the samething with partitions by resizing the partition (assuming it's at the end of the disk) that hosts the PV but I didn't have any extensive experience with resizing partitions. I have had past issues with changes to the partition table requiring reboots so I shyed away from really exploring this too deeply based on the amount of time I had. Altering the partition table on a live system also gives me and other system admins here the creeps :-). Chad ________________________________ From: "Speagle, Andy" To: Discussion list about Kickstart Sent: Tuesday, May 10, 2011 12:36 PM Subject: RE: Use entire disk as PV. > Speagle, Andy wrote: > Sent: 10 May 2011 13:53 > > > Did you use a double dash for ?--preexisting?? I didn?t notice as I > > was typing it that Outlook had > > > ?helpfully? changed the two dashes to a single emdash. Did the > > pvcreate command actually work: does that > > > disk now have a full-disk PV on it? In the pvcreate documentation, > > > I > > just came across > > > > Aye, I had used the --useexisting option with the double-hyphen. > > > > > Here?s something else to try: > > > > > > echo "part pv.01 --onpart sdb --grow" >> /tmp/custom-part echo > > > "volgroup rootvg /dev/sdb pv.01" >> /tmp/custom-part > > > > > > If kickstart refuses to create a partition on sdb, you will have to > > do something like > > > > > > lvm pvcreate /dev/sdb > > > lvm vgcreate rootvg /dev/sdb > > > echo "logvol / --fstype ext3 --name=root --vgname=rootvg --size=4096 > > --useexisting" >> /tmp/custom-part > > > > This seems to have worked for the RHEL 6 installer. RHEL 5.5 still > > barfs on this saying that no VG exists with the name "rootvg" ... > > > > I have filed a BZ with this as a feature request. I hope this gets > > added at some point. It should be fairly trivial to do. > > Is it any better in RHEL 5.6? > Were you able to tell if the lvm commands in the %pre script actually > worked? Try them manually from a kickstart shell; that may give a clue > where things are going wrong. If the volgroup is being created, maybe > kickstart in RHEL 5.5 isn't re-reading the disk information. Maybe > there's some way to get it to do that, or at worst get your %pre script to > detect whether the disk has volgroup information. If not, create it and > reboot; else write /tmp/custom-part to create the logical volumes and > continue the install. RHEL 5.6 suffers the same fate. I received a response on my BZ for this feature request that native support for PVs on entire disk will be included for anaconda somewhere in RHEL 6 ... but that it will not be back-ported to any RHEL 5 anaconda. All of the LVM commands are working just duckily. Anaconda just refuses to handle this situation. The error is quite explicit that it's an issue parsing the kickstart file. Anaconda is expecting a "volgroup" definition and won't continue without one. -Andy _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list -------------- next part -------------- An HTML attachment was scrubbed... URL: From beyonddc.storage at gmail.com Fri May 20 17:32:23 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Fri, 20 May 2011 13:32:23 -0400 Subject: Specifying the kickstart file to use a repo on a USB device Message-ID: Hi, I have a question. Is there a way to specify in the kickstart file to use a repo on a USB device? I did this on my kickstart file but no luck "repo --name=my_repo --baseurl=hd:sdb1/MY_REPO" On screen, I see the following warning message.... "YumRepo Warning: not using ftp, http[s], or file for repos, skipping - hd:sdb1/MY_REPO" What would be the correct syntax to point to a USB device? Thanks!! - David -------------- next part -------------- An HTML attachment was scrubbed... URL: From beyonddc.storage at gmail.com Fri May 20 18:13:22 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Fri, 20 May 2011 14:13:22 -0400 Subject: Specifying the kickstart file to use a repo on a USB device In-Reply-To: References: Message-ID: I think I got it working. What I did is in the %pre, I mount my USB drive first and then I set the repo using --baseurl=file:/tmp/my_repo Let me know if there's any other way to point to a repo on an USB device in the kickstart file. Thanks, David On Fri, May 20, 2011 at 1:32 PM, Chun Tat David Chu < beyonddc.storage at gmail.com> wrote: > Hi, > > I have a question. Is there a way to specify in the kickstart file to use > a repo on a USB device? > > I did this on my kickstart file but no luck > "repo --name=my_repo --baseurl=hd:sdb1/MY_REPO" > > On screen, I see the following warning message.... > "YumRepo Warning: not using ftp, http[s], or file for repos, skipping - > hd:sdb1/MY_REPO" > > What would be the correct syntax to point to a USB device? > > Thanks!! > > - David > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From clumens at redhat.com Fri May 20 19:08:09 2011 From: clumens at redhat.com (Chris Lumens) Date: Fri, 20 May 2011 15:08:09 -0400 Subject: Specifying the kickstart file to use a repo on a USB device In-Reply-To: References: Message-ID: <20110520190809.GV18330@exeter.usersys.redhat.com> > I think I got it working. > > What I did is in the %pre, I mount my USB drive first and then I set the > repo using --baseurl=file:/tmp/my_repo > > Let me know if there's any other way to point to a repo on an USB device in > the kickstart file. What you're doing is what I would recommend. - Chris From beyonddc.storage at gmail.com Fri May 20 19:30:00 2011 From: beyonddc.storage at gmail.com (Chun Tat David Chu) Date: Fri, 20 May 2011 15:30:00 -0400 Subject: Specifying the kickstart file to use a repo on a USB device In-Reply-To: <20110520190809.GV18330@exeter.usersys.redhat.com> References: <20110520190809.GV18330@exeter.usersys.redhat.com> Message-ID: Hi Chris, Got it. Thanks for your feedback - David On Fri, May 20, 2011 at 3:08 PM, Chris Lumens wrote: > > I think I got it working. > > > > What I did is in the %pre, I mount my USB drive first and then I set the > > repo using --baseurl=file:/tmp/my_repo > > > > Let me know if there's any other way to point to a repo on an USB device > in > > the kickstart file. > > What you're doing is what I would recommend. > > - Chris > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From droidery at gmail.com Tue May 31 13:26:09 2011 From: droidery at gmail.com (Amit Saha) Date: Tue, 31 May 2011 09:26:09 -0400 Subject: Installer problem with a custom Fedora Spin Message-ID: Hello all: Just been playing around with creating a custom Fedora spin using the live cd tools. In my kickstart file [1] I have specified a: "part / --size 10000", since I am installing a lot of custom packages. (Actually 5G should be enough). But anyway, so, when I start the installer (/usr/bin/liveinst) from the LiveCD environment, it crashes with: Traceback (most recent call last): File "/usr/sbin/anaconda", line 507, in from pyanaconda import kickstart File "/usr/lib/python2.7/site-packages/pyanaconda/kickstart.py", line 21, in from storage.deviceaction import * File "/usr/lib/python2.7/site-packages/pyanaconda/storage/__init__.py", line 31, in import parted File "/usr/lib/python2.7/site-packages/parted/__init__.py", line 53, in from partition import Partition File "/usr/lib/python2.7/site-packages/parted/partition.py", line 249, in partitionFlag[__flag] = _ped.partition_flag_get_name(__flag) ValueError: Invalid flag provided. I am not quite sure, what is going wrong here? Any ideas/comments appreciated. [1] https://bitbucket.org/amitksaha/custom_linux/src/a671b173396a/fedora-live-scientific.ks Many thanks in advance. -Amit -------------- next part -------------- An HTML attachment was scrubbed... URL: