[ngw] 1 year reduce job

Joe Brugaletta JBrugaletta at braytonlaw.com
Tue Nov 7 17:08:01 UTC 2017


I think it used swap at some point because:

Mem:	 10021M total,	 9620M used,	  401M free,       12M buffers
Swap:	 4102M total,	 1469M used,	 2632M free,	 5827M cached

... but if I do "vmstat 2" it shows "si" and "so" as 0 for the most part.. (swap in/out).




>>> "James Taylor" <James.Taylor at eastcobbgroup.com> 11/3/2017 3:16 PM >>>
You can run top to get an idea of process usage.
cat /proc/meminfo will tell you whether you're using any swap.
swap usage is a pretty sure sign you're out of RAM. You shouldn't be using any to speak of.
-jt

James Taylor
678-697-9420
james.taylor at eastcobbgroup.com



>>> "Joe Brugaletta" <JBrugaletta at braytonlaw.com> 11/3/2017 6:00 PM >>> 
Possibly.. but I'm by no means a linux guru. Any tips on figuring that stuff out? Disk should be good. flash arrays, etc. 





>>> "James Taylor" <James.Taylor at eastcobbgroup.com> 11/3/2017 2:48 PM >>>
Check memory and disk utilization?
You may be thrashing. GW14 uses a lot more memory than GW12 at rest, so you may be exceeding thresholds you weren't hitting previously.
Deleting mail is way more intensive than any other GW process, so it would show up there first.
-jt

James Taylor
678-697-9420
james.taylor at eastcobbgroup.com



>>> "Joe Brugaletta" <JBrugaletta at braytonlaw.com> 11/3/2017 5:23 PM >>> 
I used to be able to run these jobs during the day, no one would notice, etc.. but since gw14 maybe.. when I try to purge emails over 1 year old.. it'll lock user db's, etc for a long time.. users get "unable to open post office files" and such. End up having to kill the jobs. any tips?







_______________________________________________
ngw mailing list
ngw at ngwlist.com
http://ngwlist.com/mailman/listinfo/ngw





_______________________________________________
ngw mailing list
ngw at ngwlist.com
http://ngwlist.com/mailman/listinfo/ngw




More information about the ngw mailing list