Weekly Samba & Mysql stress testing

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Weekly Samba & Mysql stress testing

Rick Gonzalez
December 19, 2005

changeset:   8269:ac3ceb2d37d1
tag:         tip
user:        [hidden email]
date:        Wed Dec  7 09:44:38 2005 +0000
summary:     Fix the startup behavior when the tdb file exists.

* Testing Hardware:

    HW ID   Model   Procs.   Mem.    ARCH     Distro
    -----   -----   ------   ----   ------    ------
    A       x335    2        2GB    x86_32  SLES9-GA-SP2
    B       HS20    2        2GB    x86_64  SLES9-GA-SP2
    C       HS20    2        1GB    x86_64  SLES9-GA-SP2

* Xen Setups(NOTE: for HW ID look above):
                                         Dom0
    DomUs(same config)
    Setup
    ID    HW ID   CPUs    Mem    DomUs  CPUs  Mem  PAE
    ---   -----   -----   ----   -----  ----  ---  ---
    1       A       1     256MB   2      1   512MB NO
    2       B       1     256MB   2      1   512MB NO
    3       C       1     256MB   2      1   256MB NO
    4       A       1     256MB   2      1   512MB YES

* Test Status(NOTE: for Setup ID look above):

    - All stress testing is done DomU(client) to DomU(server).

    Setup ID    Workload   Test Tool   24hr   96hr   Errors(Y/N)
    --------   ---------   ---------   ----   ----   -----------
    1           MySql       DOTS        P      P         N
    2           MySql       DOTS        P      P         N
    3           Samba       fsstress    F      F         Y
-----------------------------------------------------------------

        Legend: P=pass, F=fail, Y=yes, N=no

* Issues / Comments:

Setup ID #3:

- After working with Anthony Liguori and Steven French we found out why
Samba was not running. I had a cifs unsupported option in smb.conf.

- I kicked off a 96hr run on the Samba setup and the run lasted for
about 14hrs only. Here are the reasons:

** Server Side(Samba) cat /var/log/samba/:

Many messages like the following:
   ERROR: string overflow by 3 in safe_strcat
[p0/d0XXXXXXXX/d19XX/d1fXXXXXXXXXXXXXXXX/d146XXXXXX]


** Client Side(fsstress) dmesg:

oom-killer: gfp_mask=0x1d2
DMA per-cpu:
cpu 0 hot: low 62, high 186, batch 31
cpu 0 cold: low 0, high 62, batch 31
Normal per-cpu: empty
HighMem per-cpu: empty

Free pages:        2176kB (0kB HighMem)
Active:27438 inactive:27301 dirty:0 writeback:0 unstable:0 free:544
slab:5688 mapped:54726 pagetables:1004
DMA free:2176kB min:2076kB low:2592kB high:3112kB active:109752kB
inactive:109204kB present:270336kB pages_scanned:198999
all_unreclaimable? no
lowmem_reserve[]: 0 0 0
Normal free:0kB min:0kB low:0kB high:0kB active:0kB inactive:0kB
present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
HighMem free:0kB min:128kB low:160kB high:192kB active:0kB inactive:0kB
present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
DMA: 2*4kB 13*8kB 1*16kB 2*32kB 1*64kB 1*128kB 1*256kB 1*512kB 1*1024kB
0*2048kB 0*4096kB = 2176kB
Normal: empty
HighMem: empty
Swap cache: add 293072, delete 293072, find 20825/21244, race 0+0
Free swap  = 0kB
Total swap = 1156640kB
Out of Memory: Killed process 20830 (fsstress).
oom-killer: gfp_mask=0x1d2
DMA per-cpu:
cpu 0 hot: low 62, high 186, batch 31
cpu 0 cold: low 0, high 62, batch 31
Normal per-cpu: empty
HighMem per-cpu: empty

Free pages:        2088kB (0kB HighMem)
Active:27811 inactive:27787 dirty:0 writeback:0 unstable:0 free:522
slab:4830 mapped:55681 pagetables:1066
DMA free:2088kB min:2076kB low:2592kB high:3112kB active:111244kB
inactive:111148kB present:270336kB pages_scanned:342652
all_unreclaimable? yes
lowmem_reserve[]: 0 0 0
Normal free:0kB min:0kB low:0kB high:0kB active:0kB inactive:0kB
present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
HighMem free:0kB min:128kB low:160kB high:192kB active:0kB inactive:0kB
present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
DMA: 32*4kB 1*8kB 0*16kB 1*32kB 0*64kB 1*128kB 1*256kB 1*512kB 1*1024kB
0*2048kB 0*4096kB = 2088kB
Normal: empty
HighMem: empty
Swap cache: add 533307, delete 533307, find 123324/124022, race 1+0
Free swap  = 0kB
Total swap = 1156640kB
Out of Memory: Killed process 17448 (find).
oom-killer: gfp_mask=0x1d2
DMA per-cpu:
cpu 0 hot: low 62, high 186, batch 31
cpu 0 cold: low 0, high 62, batch 31
Normal per-cpu: empty
HighMem per-cpu: empty

Free pages:        1980kB (0kB HighMem)
Active:27679 inactive:27879 dirty:0 writeback:0 unstable:0 free:495
slab:4820 mapped:55609 pagetables:1062
DMA free:1980kB min:2076kB low:2592kB high:3112kB active:110716kB
inactive:111516kB present:270336kB pages_scanned:345740
all_unreclaimable? yes
lowmem_reserve[]: 0 0 0
Normal free:0kB min:0kB low:0kB high:0kB active:0kB inactive:0kB
present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
HighMem free:0kB min:128kB low:160kB high:192kB active:0kB inactive:0kB
present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
DMA: 3*4kB 2*8kB 0*16kB 1*32kB 0*64kB 1*128kB 1*256kB 1*512kB 1*1024kB
0*2048kB 0*4096kB = 1980kB
Normal: empty
HighMem: empty
Swap cache: add 533316, delete 533316, find 123324/124023, race 1+0
Free swap  = 0kB
Total swap = 1156640kB
Out of Memory: Killed process 20831 (fsstress).


- This looks like there is a memory leak caused by fsstress.
- I am going to try using another stress test.


* Results:

- Please let me know if you would like to see the results.







_______________________________________________
Xen-devel mailing list
[hidden email]
http://lists.xensource.com/xen-devel