On Tuesday, March 10, 2020 at 11:49:58 AM UTC, maiski wrote: > > Quoting Ulrich Windl <ulric...@rz.uni-regensburg.de <javascript:>>: > > For some reason I have a "watch -n30 lvs" running in a big terminal. > > On one of the op lines I see the usage of the thin pool. Of course > > this only helps before the problem... > > > > But I thought some app is monitoring the VG; wasn't there some space > > warning before the actual problem? > > Of course there was. But atm of failure there was none visible, which > does not excuse that beforehand I had created 3 new and downloaded a > minimal template for fun, so... why can it be simple, when it can be > complicated >
Qubes 4.1 (in development) has added a warning (in addition to the current lvm space usage warning) for lvm metadata usage above a threshold. 4.0 doesn't have the metadata nearing full warning, and that's what tends to cause these types of thinpool issues. In addition to the warning, Qubes 4.1 is also doubling (vs. the lvm default value) the amount of space set aside for lvm thinpool metadata which will substantially reduce the chances of ever hitting this issue under 4.1. Brendan PS - above is not helpful for recovering this machine, of course. However, recovery from this can be very difficult and even after recovery not guaranteed to recover all the data. The Qubes devs are aware of this and very much want to avoid these issues in the next release. -- You received this message because you are subscribed to the Google Groups "qubes-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to qubes-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/qubes-users/54c72ea0-be0b-4414-bc01-7fee409b7c73%40googlegroups.com.
