Re: Strange performance issues under CentOS 5.1

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]



On Feb 12, 2008, at 21:57, William L. Maltby wrote:

Check BIOS settings? For memory, CAS etc. the same? Disk hardware the
same and specified identically?

Pretty much all the same. They are standard Lenovo desktops, with a 3.4 GHz Core 2 Duo and 3 GB of memory (the BIOS doesn't let the OS address more than 3 GB).

Presumming that nothing is found there, install system accounting
packages and run some SAR reports. You may see a clue in them.

I will try that next.

Any "tweaks" on the old system you forgot to apply on the new? Elevator,
buffer flush interval changes, etc?

Nope, it's a vanilla kickstart install. I am using the same scripts (with only slight variations) to build both the 4.x and 5.x systems. I didn't do anything to "tune" the 4.x systems. That's one of the things I like about CentOS (and RHEL for that matter): they just work out of the box. A build (make/gcc) is typically CPU bound. I don't understand how it can get only 11% of the CPU. That's unheard of.

Any other noticeable things on there that may cause it? Presume the
slowdown is caused by a process that you are not looking at. "Hangs"
while some other process is waiting or tying up the CPU. Try running
top.

There is nothing else running on the systems at the time. I have lots of terminal windows open on the 4.x system, but I am ssh'ed into the 5.x system and nobody else is logged in. If anything, the 4.x system should be slower since there are many more things running on it.

I notice an execve shows on the new one that is not in the old. One says
"hmmm....".

The execve is in the other as well, just a few lines further down (i.e., it didn't make the cut when I did the "head strace.log").

What does swapon -s show?

CentOS 4.6:
Filename Type Size Used Priority /dev/mapper/VolGroup00-LogVol00 partition 4095992 210592 -1

CentOS 5.1:
Filename Type Size Used Priority /dev/mapper/VolGroup00-LogVol00 partition 3047416 120 -1

Is the system "seeing" the same amount of memory "available" or have
BIOS settings in one reduced available?

CentOS 4.6:
    MemTotal:      3113612 kB
    MemFree:         24868 kB
    Buffers:         23164 kB
    Cached:        2428508 kB
    SwapCached:     136108 kB
    Active:        1685808 kB
    Inactive:      1257004 kB
    HighTotal:     2226560 kB
    HighFree:         1088 kB
    LowTotal:       887052 kB

CentOS 5.1:
    MemTotal:      3114452 kB
    MemFree:       2722528 kB
    Buffers:         84592 kB
    Cached:         174804 kB
    SwapCached:          0 kB
    Active:         187084 kB
    Inactive:       137220 kB
    HighTotal:     2226560 kB
    HighFree:      1972860 kB
    LowTotal:       887892 kB

If all new equipment on the new one, open her up and reseat all
connections, PCI cards and mem sticks. Make sure all power connectors
are well seated to MB and drives.

I will double check that. I may also try to install 4.6 on the HW that is currently running 5.1 and see if there is a problem. With my kickstart scripts I can have the system up and running in less than 30 minutes. That is probably the best next step to rule out any HW issues. I'll do that as soon as I get into the office.

Front side bus and memory speeds set the same in BIOS?

Should be.

That's all I can think of that may be even remotely related ATM

Thanks for all the excellent suggestions, Bill.

Alfred

_______________________________________________
CentOS mailing list
CentOS@xxxxxxxxxx
http://lists.centos.org/mailman/listinfo/centos

[Index of Archives]     [CentOS]     [CentOS Announce]     [CentOS Development]     [CentOS ARM Devel]     [CentOS Docs]     [CentOS Virtualization]     [Carrier Grade Linux]     [Linux Media]     [Asterisk]     [DCCP]     [Netdev]     [Xorg]     [Linux USB]
  Powered by Linux