Re: [Qemu-block] Intermittent failure of iotest 203

2018-03-05 Thread Stefan Hajnoczi
On Fri, Feb 23, 2018 at 02:03:15PM +0100, Max Reitz wrote:
> Hi,
> 
> iotest 203 relatively often fails for me, at least when run in parallel.
>  When I run the following concurrently on four shells:
> 
> $ while TEST_DIR=/tmp/t1 ./check -T -qcow2 203; do; done
> $ while TEST_DIR=/tmp/t2 ./check -T -qcow2 203; do; done
> $ while TEST_DIR=/tmp/t3 ./check -T -qcow2 203; do; done
> $ while TEST_DIR=/tmp/t4 ./check -T -qcow2 203; do; done
> 
> Very quickly (like under ten iterations), at least one of those starts
> to hang and then fails because of a timeout in vm.get_qmp_event().
> 
> Before digging deeper into the ppoll() dungeon* myself, I decided to
> report this so I wouldn't have to. :-)

Thanks for reporting it.  I will send a patch and CC you.

Stefan

> *Backtrace:
> 
> (gdb) bt
> #0  0x7f354137b4d6 in ppoll () at /lib64/libc.so.6
> #1  0x55b659144299 in ppoll (__ss=0x0, __timeout=0x7ffe4eaca230,
> __nfds=, __fds=) at
> /usr/include/bits/poll2.h:77
> #2  0x55b659144299 in qemu_poll_ns (fds=,
> nfds=, timeout=timeout@entry=39512999619000) at
> util/qemu-timer.c:334
> #3  0x55b6591450a3 in os_host_main_loop_wait (timeout= out>) at util/main-loop.c:255
> #4  0x55b6591450a3 in main_loop_wait (nonblocking=)
> at util/main-loop.c:515
> #5  0x55b658d4a253 in main_loop () at vl.c:1933
> #6  0x55b658d4a253 in main (argc=, argv= out>, envp=) at vl.c:4757
> 
> Max
> 





signature.asc
Description: PGP signature


[Qemu-block] Intermittent failure of iotest 203

2018-02-23 Thread Max Reitz
Hi,

iotest 203 relatively often fails for me, at least when run in parallel.
 When I run the following concurrently on four shells:

$ while TEST_DIR=/tmp/t1 ./check -T -qcow2 203; do; done
$ while TEST_DIR=/tmp/t2 ./check -T -qcow2 203; do; done
$ while TEST_DIR=/tmp/t3 ./check -T -qcow2 203; do; done
$ while TEST_DIR=/tmp/t4 ./check -T -qcow2 203; do; done

Very quickly (like under ten iterations), at least one of those starts
to hang and then fails because of a timeout in vm.get_qmp_event().

Before digging deeper into the ppoll() dungeon* myself, I decided to
report this so I wouldn't have to. :-)

*Backtrace:

(gdb) bt
#0  0x7f354137b4d6 in ppoll () at /lib64/libc.so.6
#1  0x55b659144299 in ppoll (__ss=0x0, __timeout=0x7ffe4eaca230,
__nfds=, __fds=) at
/usr/include/bits/poll2.h:77
#2  0x55b659144299 in qemu_poll_ns (fds=,
nfds=, timeout=timeout@entry=39512999619000) at
util/qemu-timer.c:334
#3  0x55b6591450a3 in os_host_main_loop_wait (timeout=) at util/main-loop.c:255
#4  0x55b6591450a3 in main_loop_wait (nonblocking=)
at util/main-loop.c:515
#5  0x55b658d4a253 in main_loop () at vl.c:1933
#6  0x55b658d4a253 in main (argc=, argv=, envp=) at vl.c:4757

Max



signature.asc
Description: OpenPGP digital signature