Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I have a server with 16GB, i7 and the likes. Just a while ago I tried to restart eAccellerator with faulty parameters and the Commited memory graph in munin jumped right off the screen to 1.32TB (Yes, teraBytes). The site works normaly (it will not EVER require this amount of memory) but the whole thing is both annoying and awkward.

  1. Is this potentially dangerous?
  2. How can I make it normal again ?

Following are the munin daily graph:

Committed memory jumped to 1.32TB after a faulty start of apache

And the output of meminfo

root@base6 [~]# cat /proc/meminfo
MemTotal:     16427408 kB
MemFree:        108732 kB
Buffers:        727704 kB
Cached:        7776936 kB
SwapCached:          0 kB
Active:        9493340 kB
Inactive:      4581064 kB
HighTotal:           0 kB
HighFree:            0 kB
LowTotal:     16427408 kB
LowFree:        108732 kB
SwapTotal:     2104504 kB
SwapFree:      2104372 kB
Dirty:           87988 kB
Writeback:           0 kB
AnonPages:     5570240 kB
Mapped:         181864 kB
Slab:          2148044 kB
PageTables:      35964 kB
NFS_Unstable:        0 kB
Bounce:              0 kB
CommitLimit:  10318208 kB
Committed_AS: 1418642716 kB
VmallocTotal: 34359738367 kB
VmallocUsed:    264108 kB
VmallocChunk: 34359474035 kB
HugePages_Total:     0
HugePages_Free:      0
HugePages_Rsvd:      0
Hugepagesize:     2048 kB

Any pointers are deeply appreciated!

share|improve this question
add comment

1 Answer

Try the following:

Run "top". Press "f" for adding a column and press "p" for adding the column "SWAP". After that, you can sort the display of top with "<" and ">". Then you should be able to tell exactly which processes "uses" the swap-memory.

I had jungledisk-server running and it "used" (since it's not really used but also displayed in munin) 50GB of committed RAM. I got rid of junglediskserver and will be using the backup solution from my hoster.

Hope you can get back to the normal munin graph :)

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.