for progress on the new ORES
cluster. TL;DR: hardware is installed and in the process of being setup.
Subscribe to
to track the actual
switch from SCB hardware in EQIAD.
Zppix, thanks for the note. In this case, you can see that all relevant
parties had been informed and had commented re. ongoing efforts in the phab
cards and via wikitech-l.
See:
.
org/T165171
On Fri, Jun 16, 2017 at 9:12 AM, Marko Obrovac <mobrovac(a)wikimedia.org>
wrote:
Hello zppix,
Thank you for reaching out. We are aware of resource utilisation problems
on SCB hosts. After the incident, we have limited the amount of memory the
PDF rendering service can use~[1]. However, it is important to note that
most of the time ORES is the service using most of the resources on these
machines, and they are shared amongst 9 different services. In light of
that fact, we have previously recommended commissioning a separate cluster
for ORES. As far as I know, the budget should be there and it should be
implemented in FY 2017/2018, but will let Aaron correct me here in case I'm
wrong :)
Cheers,
Marko Obrovac, PhD
Senior Services Engineer
Wikimedia Foundation
[1]
https://gerrit.wikimedia.org/r/#/c/358888/
On 16 June 2017 at 00:10, zppix e <megadev44s.mail(a)gmail.com> wrote:
Hello, I am investigating the 6-13-17 incident
involving ores as a
volunteer for wikimedia AI, I was wondering if theres a way PDFrender can
interact less server resource wise when it comes to being on the same
server as ORES to keep ORES from failing, if not is it possible that either
ORES or PDFRender get moved to different servers perhaps? If you have any
questions, comments, or concerns please reply to me here or #wikimedia-ai
thanks!
Zppix
Volunteer developer for WMF
enwp.org/User:Zppix
_______________________________________________
Services mailing list
Services(a)lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/services