11:00:13 <oneswig> #startmeeting scientific-sig
11:00:14 <openstack> Meeting started Wed Feb 26 11:00:13 2020 UTC and is due to finish in 60 minutes.  The chair is oneswig. Information about MeetBot at http://wiki.debian.org/MeetBot.
11:00:15 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
11:00:17 <openstack> The meeting name has been set to 'scientific_sig'
11:00:30 <oneswig> greetings
11:01:10 <oneswig> #link today's agenda (such as it is) https://wiki.openstack.org/wiki/Scientific_SIG#IRC_Meeting_February_26th_2020
11:02:48 <noggin143> lo
11:03:30 <noggin143> got the ironic team at CERN this week on the midcycle F2F
11:03:35 <oneswig> Hi noggin143
11:03:40 <oneswig> how is it going?
11:04:09 <verdurin> Morning.
11:04:17 <janders> g'day!
11:04:19 <oneswig> Hello verdurin janders
11:04:26 <janders> sorry finishing off some work on my 4wd still :)
11:04:45 <oneswig> hardware hacking?
11:05:23 <noggin143> and a chance to tour some of the facilities
11:05:57 <oneswig> noggin143: I joined the large-scale sig meeting this morning and suggested a user story for large-scale baremetal
11:06:27 <oneswig> There was discussion there about a session track at the Vancouver PTG
11:06:41 <ttx> Opendev actually but yes
11:06:50 <oneswig> Ah, hi ttx
11:06:55 <oneswig> thanks :-)
11:07:24 <oneswig> also a hope that the CERN team might contribute some experiences to https://etherpad.openstack.org/p/scaling-stories
11:08:00 <janders> that would be a very interesting read! noggin143 +1
11:08:04 <noggin143> will have a look...
11:08:31 <oneswig> thanks noggin143, share and enjoy!
11:11:37 <oneswig> Martial has requested half a day (or slightly more) for the scientific sig at Vancouver
11:11:52 <oneswig> I haven't seen the scheduling yet
11:15:46 <oneswig> Another useful development was that the oslo.metrics blueprint is getting a good reception - https://review.opendev.org/#/c/704733/
11:20:11 <oneswig> verdurin: are you involved with the workshop for https://cloud.ac.uk/ this year?
11:23:22 <noggin143> We'll have a couple of folk there too (Theodoros and Joao)
11:23:49 <oneswig> noggin143: great - isn't Theodoros presenting?
11:24:08 <noggin143> yes, both are
11:24:34 <oneswig> I know Bharat from our team is looking forward to meeting Theodoros
11:25:18 <noggin143> Theodoros on the current state / ongoing activities and Joao on the EU ARCHIVER project using public cloud storage for science
11:26:30 <verdurin> oneswig: Yes, I am one of the organisers, and I'm chairing a session.
11:27:07 <oneswig> It looks like it will be another good year, hope it goes smoothly verdurin
11:27:26 <verdurin> There was a bit of competition to chair the restricted data session, and I lost.
11:28:20 <oneswig> I was somewhat alarmed to follow someone from the Alan Turing Institute - was intending to present details on the paper they published on this myself...
11:28:44 <oneswig> ... I hastily redrafted my content
11:29:24 <oneswig> OK well I don't think there's much to cover this week, let's do AOB
11:29:28 <oneswig> #topic AOB
11:30:23 <oneswig> I'm getting to the bottom of doing SR-IOV on top of 2-port bonded NICs, am hoping to write up the findings within a week or two
11:31:07 <oneswig> should be interesting but the question of "was it worth the trouble" is harder to answer
11:31:24 <janders> in case of fitting the winch and radios on the 4wd - for sure! :)
11:31:29 <noggin143> we're looking into a tough ceph issue from last week, current suspicion is that LZ4 compression may have problems
11:31:46 <noggin143> but nothing for sure
11:31:54 <oneswig> noggin143: ouch.  How do you isolate that?
11:32:19 <noggin143> had to rebuild the osd-maps. Dan will write up something once the analysis is complete
11:32:35 <oneswig> sounds interesting, thanks.
11:32:48 <oneswig> Do you know how the seastar messenger work is progressing btw?
11:33:36 <noggin143> I don't know the details, was due to get an update at cephalacon but it got cancelled (South Korea not the best place to visit at the moment))
11:34:29 <oneswig> I'm having not very much success in finding the point in the chain where packets are being discarded, when it appears to be set up for end-to-end flow control.  Very niggly.
11:36:10 <oneswig> OK, any more AOB to share?
11:36:28 <janders> nothing from my side
11:36:39 <verdurin> Nor from me.
11:37:57 <oneswig> In which case I shall return to my dark cave of LAGs, SRIOV and troubleshooting packet drops...
11:38:08 <oneswig> thanks all, see you next week verdurin
11:38:11 <oneswig> #endmeeting