Re: cyrus-murder problems with database corruption in the frontend/master

[João Assad]

Henrique de Moraes Holschuh wrote:
Just thought of something.  Please set the
vm.overcommit_memory syscall to 2  (it is available in /proc/sys, I think.
But the right way is to use /etc/sysctl.conf and sysctl).
Make *really* sure you have enough swap when you do that. You will *really*
need it.
Some look on /proc/meminfo (especially on CommitLimit and Committed_AS)
might shed some light if that is the problem.
 

Yeah I thought that might help too, I did that yesterday after the last 
corruption
I also set the overcommit_ratio to 100

So far everything is running smooth. but I also recuced the 
mupdate_workers configs, which usually makes cyrus last longer.
Im getting anxious here, I already have 643M worth of straces since I 
started it.  So far no ENOMEMs .

My current meminfo
MemTotal:      2074268 kB
MemFree:        277240 kB
Buffers:         93244 kB
Cached:        1437800 kB
SwapCached:       3216 kB
Active:        1247724 kB
Inactive:       402988 kB
HighTotal:     1178756 kB
HighFree:       153984 kB
LowTotal:       895512 kB
LowFree:        123256 kB
SwapTotal:     2096472 kB
SwapFree:      2093068 kB
Dirty:            1448 kB
Writeback:           0 kB
Mapped:         136136 kB
Slab:            46308 kB
CommitLimit:   4170740 kB
Committed_AS:  1494736 kB
PageTables:      90924 kB
VmallocTotal:   114680 kB
VmallocUsed:      2072 kB
VmallocChunk:   112208 kB
HugePages_Total:     0
HugePages_Free:      0
Hugepagesize:     2048 kB

---
Cyrus Home Page: http://asg.web.cmu.edu/cyrus
Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu
List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html