linux server slow.

unix syzadmin unixsyzadmin at gmail.com
Wed Dec 21 05:48:06 UTC 2005


Hi Guys,

We finally found a fix.  I have booted Linux with "linux numa=off".  Now the
linux partition is able to see all the memory allocated for it.


[Numa is Non Uniform Memory Access, it's a new architecture developed to
overcome the bottlenecks of SMP (Symmetric Multiprocessing), that arise when
there are dozens of CPU's trying to access memory on a shared memory bus.]
Since we have only 4 CPU's we donot need the numa.

Thanks & Regards,
-GnanaShekar-


On 12/14/05, unix syzadmin <unixsyzadmin at gmail.com> wrote:
>
>
>
>   Hi,
> >
> > The configuration of the 2 LPAR's is done such that the CPU and memory
> > resources are equally divided.
> > Even though the LPAR's are allocated 3.5GB RAM each, the Linux top or
> > free command shows only 360MB as the available RAM.
> >
> >  RHEL AS4
> > ========
> > [root at bangpcplin3 ~]# free -m
> >              total       used       free     shared    buffers
> > cached
> > Mem:           360        352          7          0          5
> > 74
> > -/+ buffers/cache:        271         88
> > Swap:         3074        319       2755
> >
> > Have anybody else faced a similar circumstance, where the available
> > memory is more than what is seen by the linux operating system? if so please
> > guide me.
> >
> > Thanks & Regards,
> > -GnanaShekar-
> >
> >
> >  On 12/14/05, Larry D Sorensen <larry.sorensen at juno.com > wrote:
> > >
> > > It looks like your problem may be with the Virtual IO partition rather
> > > than with your Linux partition. I would start there and look back over
> > >
> > > your configurations for that. IBM has some good documentation on LPARs
> > > and VIO on their web site.
> > >
> > > On Wed, 14 Dec 2005 09:27:32 +0530 unix syzadmin <unixsyzadmin at gmail.com
> > > >
> > > writes:
> > > > Hi,
> > > >
> > > > We have setup 2 LPARs (Logical Partitions) on IBM p5-550.
> > > > One for AIX 5.3, The 2nd one for the RHEL AS4.
> > > > The server has 4 CPU's and 8GB RAM.  It has two 140GB hard disk.
> > > > The 2 LPAR's are configured with a virtual ethernet client and
> > > > virtual
> > > > SCSI client.
> > > >
> > > > We have created 2 volume groups and each volume group containing 2
> > > > logical
> > > > volumes.  We have assiged 2 logical volumes to each virtual host.
> > > >
> > > > The CPU and memory is divided as follows
> > > > VIO partition -->.5 (5 Virtual processors),  512MB RAM
> > > > AIX partition --> 1.7 (17 Virtual processors), 3.5GB RAM
> > > > RedHat AS4 partition --> 1.7 (17 Virtual processors), 3.5GB RAM
> > > >
> > > > Just see a sample of output of ping to another server in the network
> > > > from
> > > > RHEL AS4.
> > > >
> > > > [root at bangpcplin1 ~]# ping bangpcplin3
> > > > PING bangpcplin3.bea.com ( 172.22.43.47) 56(84) bytes of data.
> > > > 64 bytes from bangpcplin3.bea.com ( 172.22.43.47): icmp_seq=0 ttl=64
> > > > time=
> > > > 21.8 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=1 ttl=64
> > > > time=
> > > > 9.05 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=2 ttl=64
> > > > time=
> > > > 8.97 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=3 ttl=64
> > > > time=
> > > > 10.0 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=4 ttl=64
> > > > time=
> > > > 13.0 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47 ): icmp_seq=5 ttl=64
> > > > time=
> > > > 11.0 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=6 ttl=64
> > > > time=
> > > > 8.98 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=7 ttl=64
> > > > time=
> > > > 6.60 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=8 ttl=64
> > > > time=
> > > > 10.9 ms
> > > > 64 bytes from bangpcplin3.bea.com ( 172.22.43.47): icmp_seq=9 ttl=64
> > > > time=
> > > > 9.98 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=10 ttl=64
> > > > time=
> > > > 10.0 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=11 ttl=64
> > > > time=
> > > > 7.92 ms
> > > > 64 bytes from bangpcplin3.bea.com (172.22.43.47): icmp_seq=12 ttl=64
> > > > time=
> > > > 9.98 ms
> > > >
> > > > --- bangpcplin3.bea.com ping statistics ---
> > > > 13 packets transmitted, 13 received, 0% packet loss, time 12088ms
> > > > rtt min/avg/max/mdev = 6.605/10.650/21.824/3.557 ms, pipe 2
> > > > [root at bangpcplin1 ~]#
> > > >
> > > > Also see a sample output of "iostat 5"
> > > >
> > > > [root at bangpcplin1 ~]# iostat 5
> > > > Linux 2.6.9-5.EL (bangpcplin1)  12/14/2005
> > > >
> > > > avg-cpu:  %user   %nice    %sys %iowait   %idle
> > > >            0.82    0.00    0.43    1.13   97.62
> > > >
> > > > Device:            tps   Blk_read/s   Blk_wrtn/s   Blk_read
> > > > Blk_wrtn
> > > > sda              10.61       140.42        63.16  142819579
> > > > 64242894
> > > > sdb               1.83        35.94         25.10   36551596
> > > > 25528376
> > > >
> > > > avg-cpu:  %user   %nice    %sys %iowait   %idle
> > > >            0.03    0.00    2.69    3.08   94.20
> > > >
> > > > Device:            tps   Blk_read/s   Blk_wrtn/s   Blk_read
> > > > Blk_wrtn
> > > > sda             102.19      1011.95       309.16       5080
> > > > 1552
> > > > sdb              36.25       290.04         0.00       1456
> > > > 0
> > > >
> > > > avg-cpu:  %user   %nice    %sys %iowait   %idle
> > > >            0.12    0.00    2.78    4.01   93.09
> > > >
> > > > Device:            tps   Blk_read/s   Blk_wrtn/s   Blk_read
> > > > Blk_wrtn
> > > > sda              92.79      1463.73         1.60       7304
> > > > 8
> > > > sdb              29.86       235.67         4.81       1176
> > > > 24
> > > >
> > > > avg-cpu:  %user   %nice    %sys %iowait   %idle
> > > >            0.02    0.00    2.60    3.88   93.50
> > > >
> > > > Device:            tps   Blk_read/s   Blk_wrtn/s   Blk_read
> > > > Blk_wrtn
> > > > sda             128.80      2056.00         0.00      10280
> > > > 0
> > > > sdb               0.40         0.00         4.80          0
> > > > 24
> > > >
> > > > avg-cpu:  %user   %nice    %sys %iowait   %idle
> > > >            0.04    0.00    2.79    3.35   93.83
> > > >
> > > > Device:            tps   Blk_read/s   Blk_wrtn/s   Blk_read
> > > > Blk_wrtn
> > > > sda             162.28      1534.53       621.16       7688
> > > > 3112
> > > > sdb               0.00         0.00         0.00          0
> > > > 0
> > > >
> > > >
> > > > Thanks & Regards,
> > > >
> > > > -GnanaShekar-
> > > >
> > > >
> > > >
> > > > On 12/14/05, Smith, Albert < Albert.Smith at genexservices.com> wrote:
> > > > >
> > > > > Without knowing more about the configuration of your system and of
> > >
> > > > the
> > > > > application there is very little go on. Can you provide more
> > > > details?
> > > > >
> > > > > Albert Smith
> > > > > Sr. Unix Systems Administrator
> > > > > HPCSA, RHCT
> > > > > Genex Services
> > > > > 440 E. Swedesford Rd.
> > > > > Wayne, PA 19087
> > > > > albert.smith at genexservices.com
> > > > > (610) 964-5154
> > > > >
> > > > >
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: redhat-list-bounces at redhat.com
> > > > > > [mailto:redhat-list-bounces at redhat.com] On Behalf Of unix
> > > > syzadmin
> > > > > > Sent: Tuesday, December 13, 2005 8:54 PM
> > > > > > To: redhat-list at redhat.com
> > > > > > Subject: linux server slow.
> > > > > >
> > > > > > Hi,
> > > > > >
> > > > > > The users in my team are running tests for weblogic on linux
> > > > > > (RHEL AS4).  We generally run test for weblogic on different
> > > > > > operating system and certify it.
> > > > > > They have complained that the linux server is slow.  A test
> > > > > > that usually would take 30 mins is taking about 5 hours.
> > > > > > They have not faced this kind of slowness with any other
> > > > > > server (solaris, aix & hp-unix), even the earlier servers
> > > > > > that were running linux.
> > > > > >
> > > > > > I am new to troubleshooting performance problems.  Please
> > > > suggest.
> > > > > >
> > > > > > Thanks & Regards,
> > > > > > -GnanaShekar-
> > > > > > --
> > > > > > redhat-list mailing list
> > > > > > unsubscribe
> > > > mailto: redhat-list-request at redhat.com?subject=subscribe
> > > > > > https://www.redhat.com/mailman/listinfo/redhat-list
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > > --
> > > > > redhat-list mailing list
> > > > > unsubscribe
> > > > mailto: redhat-list-request at redhat.com?subject=unsubscribe
> > > > > https://www.redhat.com/mailman/listinfo/redhat-list
> > > > >
> > > > --
> > > > redhat-list mailing list
> > > > unsubscribe
> > > > mailto: redhat-list-request at redhat.com ?subject=unsubscribe
> > > > https://www.redhat.com/mailman/listinfo/redhat-list
> > > >
> > > >
> > >
> > > --
> > > redhat-list mailing list
> > > unsubscribe mailto:redhat-list-request at redhat.com ?subject=unsubscribe
> > > https://www.redhat.com/mailman/listinfo/redhat-list
> > >
> >
> >
>



More information about the redhat-list mailing list