[haiku-sysadmin] Re: HDS; Outage w/ Memory Exhaustion
- From: Axel Dörfler <axeld@xxxxxxxxxxxxxxxx>
- To: haiku-sysadmin@xxxxxxxxxxxxx
- Date: Thu, 27 Oct 2016 09:43:46 +0200
Hi Andrew,
Am 27/10/2016 um 07:50 schrieb Andrew Lindesay:
It looks like HDS has been taken-out because of memory-exhaustion on the
'vmrepo' host again. It seems like 'nscd' is to blame using something
like 40% of the memory;
```
644 nscd 20 0 2828776 0.991g 396 S 0.000 40.56 10:21.55
nscd
```
[...]
I assume there's nobody with time available to re-configure the host
with more memory / swap? I guess it would be difficult to look at other
hosting options too because it would get too complex to deal with
providing access to it etc...?
Ordering more memory for the server is possible; it's currently at 16GB,
and can be upgraded to 32 GB for a one time fee of 17.65 EUR, and a
monthly fee of about 5.50 EUR.
However, given that we have an actual culprit, buying more memory
doesn't sound like the preferred solution; I have changed the nscd
configuration to restart every two hours [1].
Bye,
Axel.
[1]
http://serverfault.com/questions/463648/why-would-nscd-use-a-huge-amount-of-ram
Other related posts:
