On 09/07/2017 01:05 AM, Pierre-Yves Chibon wrote:
On Wed, Sep 06, 2017 at 05:44:41PM -0400, Stephen John Smoogen
wrote:
> Thursday 2017-09-07 the Fedora Infrastructure team will have a meeting
> in #fedora-meeting at 18:00 UTC. The agenda currently looks as follows
> on the gobby page. Please add or update as needed.
>
> = Introduction =
> We will use it over the week before the meeting to gather status and info and
> discussion items and so forth, then use it in the irc meeting to transfer
> information to the meetbot logs.
>
> = Meeting start stuff =
>
> #startmeeting Infrastructure (2017-09-07)
> #meetingname infrastructure
> #topic aloha
> #chair smooge relrod nirik abadger1999 dgilmore threebean pingou
> puiterwijk pbrobinson
I think we need to add maxamillion to this list now :)
> = Let new people say hello =
>
> #topic New folks introductions
>
> = Status / Information / Trivia / Announcements =
>
> (We put things here we want others on the team to know, but don't need
> to discuss)
> (Please use #info <the thing> - your name)
>
> #topic announcements and information
> #info PHX2 Colo Trip, Dec 4th - 9th
> #info beta freeze started 2017-09-05 - everyone
> #info Moving .stg. ips ongoing
> #info DNS breakage happened earlier this week
> #info NFS slowdowns moved to high IO problem.
Hm, not sure I understand this last statement on its own.
We are no longer seeing nfs/netapp storage be slow, instead there is
something taking up tons and tons of read BW on it.
(something like 900MB/sec since the 31st). While that isn't being slow
for us, the storage folks want it to stop so we don't mess up
performance for other groups using the same head units. :)
I am not sure whats causing it... smooge and I looked and I found one
problem on bodhi-backend01 (it was mounting /pub twice, once with
NFS4.1). Fixing that didn't change anything. Smooge saw that the
kojipkgs machines were using a fair bit of BW, but thats somewhat
"normal" for them and varnish should be caching it away from nfs.
kevin