[SOLVED] Temporary Outage - ovirt.org Website

[Michael Scherer]

Le lundi 07 mars 2016 à 14:13 +0100, Mikey Ariel a écrit :
> Hi folks,
> 
> It seems that our logs on OpenShift filled up and the ovirt.org website
> was unavailable for a few hours this morning.
> 
> The issue was resolved and the site is back online.

So to complete what Mikey said:
- we have increased the partition from 1G to 3G (this was done
automatically with previous openshift version, so insert my usual rant
about changes in *aaS), so we would have more room next time.

- a RCA (root cause analysis) show that we got 250 000 requests today
from 1 single IP at OVH, around 100 000 from bing bot reindexing the
website (on 4 or 5 ip ), and 30 000 requests from a ip of  bezeqint.net,
with a nagios check_http signature. 

While we can't do much for the first or the 2nd (besides blocking), I
would suggest that the 3rd ip owner decide to reduce the frequency of
check from 1 every 2 seconds to 1 every 5 minutes.

-- 
Michael Scherer
Sysadmin, Community Infrastructure and Platform, OSAS


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part
URL: <http://lists.ovirt.org/pipermail/infra/attachments/20160307/115e27b1/attachment.sig>