| 02:28.36 | *** join/#fredlug plarsen (n=plarsen@72-255-115-78.client.stsn.net) |
| 03:32.43 | *** join/#fredlug nombyte (n=nomb@c-71-62-193-105.hsd1.va.comcast.net) |
| 14:19.47 | *** join/#fredlug jsmith (i=jsmith@nat/digium/x-22e2b49f5f69c3b7) |
| 14:31.52 | *** join/#fredlug plarsen (n=plarsen@208.176.91.226.ptr.us.xo.net) |
| 14:32.05 | plarsen | Good morning gents! |
| 14:36.39 | jsmith | Buenos dias! |
| 14:36.41 | jsmith | Good morning! |
| 14:36.48 | jsmith | Buon giorno |
| 14:36.53 | jsmith | (or however you spell that) |
| 14:36.55 | jsmith | Bonjour! |
| 14:37.21 | jsmith | stickster: Well, you and your cohorts twisted my arm... |
| 14:37.37 | plarsen | ohhhh - I see a fight coming one! |
| 14:37.40 | plarsen | on* |
| 15:04.12 | *** join/#fredlug jsmith (i=jsmith@nat/digium/x-864e584ba312d501) |
| 15:09.59 | stickster | jsmith could whip me with one arm tied to his Asterisk server |
| 15:10.32 | jsmith | stickster: Oh, whatever... you'd tie me in knots with your excellent use of English. |
| 15:10.52 | stickster | "That is the sort of English up with which I will not put!" |
| 15:14.38 | jsmith | You make your point as eloquently as ever... |
| 15:38.27 | plarsen | lol - since the dimise of the dual, fights between gentlemen just aren't the same ;) |
| 15:38.40 | plarsen | duel* |
| 15:45.49 | jsmith | plarsen: And that's a bad thing? |
| 16:01.51 | plarsen | If what you two presented in "cursing eachother out" is an example, yes !! :) |
| 16:09.23 | stickster | plarsen: Were you able to figure out why xfs died OOM? |
| 16:14.04 | *** join/#fredlug jsmith (i=jsmith@nat/digium/x-52f7cce372307ef9) |
| 16:21.31 | plarsen | stickster: it's not just xfs - it's everything. I think it's a buggy VMWare module. |
| 16:22.01 | plarsen | I reinstalled the vmware client/toolkit and it helped ... but not totally solved. |
| 16:22.27 | plarsen | VMWare has a vmmemctl module. It was loaded and being reported as "tainting" the kernel. |
| 16:22.49 | plarsen | I did a reinstall and so far two boxes seem to have heeded and have stopped random killing almost from boot. |
| 18:21.26 | *** join/#fredlug jsmith (i=jsmith@nat/digium/x-4379297447b6cc81) |
| 19:25.30 | *** join/#fredlug plarsen (n=plarsen@208.176.91.226.ptr.us.xo.net) |
| 19:43.59 | stickster | plarsen: Is that RHEL 3? |
| 19:45.18 | stickster | Could it be low memory exhaustion? Are you running the hugemem kernel? |
| 20:12.40 | plarsen | stickster: no ... |
| 20:12.45 | plarsen | it's RHES4 |
| 20:12.54 | plarsen | "standard" smp kernel |
| 20:14.16 | plarsen | <PROTECTED> |
| 20:22.55 | stickster | plarsen: OK, may not be relevant if it's 3 or 4. |
| 20:23.02 | stickster | plarsen: How much memory in the machine? |
| 20:23.20 | plarsen | 4GB |
| 20:23.34 | plarsen | Worked for months and months with 2GB without much swap in use |
| 20:23.48 | stickster | Hmm, well that doesn't seem like so much that you need to worry about hugemem then. |
| 20:24.25 | plarsen | nope |
| 20:24.54 | stickster | What is /proc/sys/vm/lower_zone_protection on that machine? |
| 20:26.33 | stickster | Some other settings that might be important: |
| 20:26.54 | stickster | overcommit memory and ratio |
| 20:27.16 | stickster | min_free_kbytes |
| 20:28.58 | stickster | This might also help: http://people.redhat.com/nhorman/papers/rhel4_vm.pdf |
| 20:41.08 | plarsen | 1002$ cat /proc/sys/vm/lower_zone_protection |
| 20:41.08 | plarsen | 0 |
| 20:41.35 | plarsen | sandbox1*sand1-/home/oracle |
| 20:41.35 | plarsen | 1004$ cat /proc/sys/vm/overcommit_memory |
| 20:41.35 | plarsen | 0 |
| 20:41.35 | plarsen | sandbox1*sand1-/home/oracle |
| 20:41.35 | plarsen | 1005$ cat /proc/sys/vm/overcommit_ratio |
| 20:41.36 | plarsen | 50 |
| 20:45.22 | stickster | plarsen: You might want to try around 250 in lower_zone_protection and see if that helps |
| 21:51.26 | *** join/#fredlug jsmith (i=jsmith@nat/digium/x-c15e8520a33b536d) |
| 21:56.46 | *** join/#fredlug nombyte (n=nomb@c-71-62-193-105.hsd1.va.comcast.net) |