sarnold | and now I'm headed out; no visible motion on the rt | 01:46 |
---|---|---|
rbasak | https://git.launchpad.net/ubuntu/+source/ceph has taken an hour to clone so far, at 92%, current rate 69 KiB/s :-/ | 11:08 |
rbasak | Is there anything we can do to speed this up please? Would a repack help? | 11:08 |
cjwatson | I'll try a repack | 11:14 |
cjwatson | Perhaps there's some other problem but I don't have time to dig further right now | 11:14 |
rbasak | Thanks | 11:27 |
rbasak | FWIW I'm still at 92%, but if you're repacking, that presumably won't help unless I restart and meanwhile I/O will presumably be worse. | 11:28 |
rbasak | FTR my ceph clone just finished at 1h45m, for comparison with next time. | 11:52 |
cjwatson | Indeed, repacks will only be effective from the next clone. It's repacked now though | 12:13 |
kkremitzki[m] | We are having similar problems with FreeCAD's Launchpad git repo which has been timing out since the beginning of this month, perhaps a repack would help? https://code.launchpad.net/~freecad-maintainers/freecad/+git/freecad | 12:30 |
cjwatson | I can try but I think we'll probably find it's deeper than that. | 12:30 |
kkremitzki[m] | Indeed, thank you. | 12:30 |
cjwatson | Oh, that's an import, almost certainly a different problem. | 12:30 |
cjwatson | But won't hurt to try a repack. | 12:30 |
cjwatson | Hm, something went quite dramatically south not long after 0800Z today. | 12:34 |
cjwatson | Massive RAM spike | 12:34 |
cjwatson | Dead cache battery on one of the host systems apparently, which can't be helping. We've opened a ticket to get that replaced | 13:20 |
cjwatson | May explain various other bits of slowness around git recently | 13:25 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!