Hi Yuri, I did not know about scrape. It looks extremely useful, thanks ! Do you have more magic items in your bag ? ;-) Cheers P.S. And thanks to John for writing scrape :-) On 18/01/2015 01:27, Yuri Weinstein wrote: > Sounds good, Loic > > Are you aware BTW about scrape tool written by John Spray? > https://github.com/jcsp/scrape > > I use it for test runs analysis often. > > Just FYI > > Thx > YuriW > > On Sat, Jan 17, 2015 at 3:27 PM, Loic Dachary <loic@xxxxxxxxxxx> wrote: >> >> >> On 17/01/2015 23:08, Yuri Weinstein wrote: >>> Hi Loic, >>> >>> I'd love to. >>> >>> Let's chat on Monday to finalize format. >>> >>> ( >>> PS: I think format would be easier to maintain like new lines enforced, e,g. >>> >>> '701575', '701576', '701582', '701590' >>> ttp://tracker.ceph.com/issues/10543 >>> FAILED assert(m_seed < old_pg_num >> >> Yes, as long as it's consistent enough to be machine readable, that works :-) >> >>> >>> PPS: Again this ticket http://tracker.ceph.com/issues/10455 would be >>> helpful in what we are discussing here. >>> >> >> I'm not sure where the output of such a parsing would go. The update redmine API is currently broken (the read API works ok) but if it was fixed the tickets could be updated indeed. >> >> Cheers >> >>> Thx >>> YuriW >>> >>> On Sat, Jan 17, 2015 at 5:07 AM, Loic Dachary <loic@xxxxxxxxxxx> wrote: >>>> Hi Yuri, >>>> >>>> It would be great if the analysis you compile daily was machine readable. For instance, in a mail you sent to ceph-qa I read >>>> >>>> '701575', '701576', '701582', '701590' - known issue >>>> http://tracker.ceph.com/issues/10543 >>>> >>>> "FAILED assert(m_seed < old_pg_num)" >>>> >>>> (duplicate of http://tracker.ceph.com/issues/10430) >>>> >>>> which could be something like: >>>> >>>> '701575', '701576', '701582', '701590': http://tracker.ceph.com/issues/10543 FAILED assert(m_seed < old_pg_num) >>>> >>>> or any other format you find easier to use consistently. >>>> >>>> The reason I ask is because it would help me write a script that associates redmine tickets to your findings, in the context of backporting. It's just an idea, not a request ;-) >>>> >>>> Cheers >>>> >>>> -- >>>> Loïc Dachary, Artisan Logiciel Libre >>>> >> >> -- >> Loïc Dachary, Artisan Logiciel Libre >> -- Loïc Dachary, Artisan Logiciel Libre
Attachment:
signature.asc
Description: OpenPGP digital signature
