[Spacewalk-list] Repo Maintenance with spacewalk-reposync

Glennie, Jonathan - 0443 - MITLL jrglenni at ll.mit.edu
Wed Jun 8 16:08:36 UTC 2011


Hello All-

 

Sorry about the double e-mail, accidentally hit send on the first one before
I was done. 

 

I've been seeing interesting oddities with my spacewalk channels syncing
over time on Spacewalk 1.3.  One issue that keeps cropping up is my
spacewalk channels will inexplicably contain more packages than the actual
upstream channels, even though when I run the repo sync command it reports
that the upstream repo has the correct number of packages.  For instance,
this morning my CentOS OS repo had 4400+ packages in it and running the
repo-sync command reported that there were 3535 packages in the upstream
repo.  Not sure how it happens, but I get multiple copies some of the
packages.  I've been deleting all of the packages in my SW repo once a week
and letting it do a complete re-sync in order to get the correct number of
packages.  Has anyone else seen this behavior?  If so, are there any better
ways to handle it?  I've set up my repo syncs to run nightly from a cron
with the format "/usr/bin/spacewalk-repo-sync --channel centos5-os-x86_64
--url http://mirror.centos.org/centos/5/os/x86_64/ --type yum".  I would
think that's ok, but maybe I should be running something slightly different,
or have better luck setting up the syncs through the web UI?

 

Another issue I get is random errors when syncing large amounts of packages,
usually after 500 or so.  An example is shown below.

 

python: ./rpmio_internal.h:502: c2f: Assertion `fd && fd->magic ==
0x04463138' failed.

ORA-24550: signal received: [si_signo=6] [si_errno=0] [si_code=-6]
[si_int=0] [si_ptr=0x200000000] [si_addr=0xf5c]

kpedbg_dmp_stack()+360<-kpeDbgCrash()+192Cannot open
/usr/lib64/oracle/11.2/client/lib/libclntsh.so.11.1 for reading: errno=24

<-000000389B9316F9Cannot open
/usr/lib64/oracle/11.2/client/lib/libclntsh.so.11.1 for reading: errno=24

<-000000389B6938C2<-__sighandler()<-gsignal()+53<-00002AF95445D660<-00002AF9
5445D660<-00002AF95445D660<-00002AF95445D660<-00002AF95445D660<-00002AF95445
D660<-00002AF95445D660<-00002AF95445D660<-00002AF95445D660<-00002AF95445D660
<-00002AF95445D660<-00002AF95445D660<-00002AF95445D660<-00002AF95445D660<-00
002AF95445D660<-00002AF95445D660<-00002AF95445D660<-00002AF95445D660<-00002A
F95445D660

Abort

 

Re-isuing the sync commands resumes from where it left off and I can
successfully download the rest of the packages, unless there are a large
number remaining in which case I hit the error again and then rinse and
repeat until they are all downloaded.  Any idea what that may be?  Could
this be what's causing my duplicate-package issues?    Thanks in advance for
any help/insight.

 

-Jon

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/spacewalk-list/attachments/20110608/59d51e10/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 5558 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/spacewalk-list/attachments/20110608/59d51e10/attachment.bin>


More information about the Spacewalk-list mailing list