Hello,
We were testing corosync 2.4.3/libqb 1.0.1-6/sbd 1.3.1/gfs2 on 4.19 and
noticed a fundamental problem with realtime priorities:
- corosync runs on CPU3, and interrupts for the NIC used by corosync are
also routed to CPU3
- corosync runs with SCHED_RR, ksoftirqd does not (should it?), but
wi
Edwin,
Hello,
We were testing corosync 2.4.3/libqb 1.0.1-6/sbd 1.3.1/gfs2 on 4.19 and
noticed a fundamental problem with realtime priorities:
- corosync runs on CPU3, and interrupts for the NIC used by corosync are
also routed to CPU3
- corosync runs with SCHED_RR, ksoftirqd does not (should it
Apologies for top posting, the strace you asked for is available here (although
running strace itself had side-effect of getting corosync out of the live lock):
https://clbin.com/9kOUM
Best regards,
--Edwin
From: Jan Friesse
Sent: 14 February 2019 18:34
T
Edvin,
Apologies for top posting, the strace you asked for is available here (although
running strace itself had side-effect of getting corosync out of the live lock):
Yep, I know, but this was after strace finished, right? So links should
contain time when corosync was stucked, right?
htt
Hi!
IMHO any process running at real-time priorities must make sure that it
consumes the CPU only for shorrt moment that are really critical to be
performed in time. Specifically having some code that performs poorly (for
various reasons) is absolutely _not_ a candidate to be run with real-time
pr
I would expect that, as strace interrupts the RT task to query the code; you
should run strace at the same RT priority ;-)
>>> Edvin Torok 14.02.19 19.54 Uhr >>>
Apologies for top posting, the strace you asked for is available here (although
running strace itself had side-effect of getting coro
Ulrich Windl napsal(a):
Hi!
IMHO any process running at real-time priorities must make sure that it
consumes the CPU only for shorrt moment that are really critical to be
performed in time. Specifically having some code that performs poorly (for
various reasons) is absolutely _not_ a candidate t