Ok, on one hand glad to hear that is an artifact, on the other hand that means the issue is still out there... I'm not a maintainer of Freeswitch, I am an user of it. However I am trying to supply the maintainers with any relevant information regarding this issue. As you mentioned, I see the socket being modified and passed around in the source code. Do you think the implementation is OK? I do not know if that's standard practice when consuming libpq. 2018-03-02 17:36 GMT-03:00 Andres Freund <andres@xxxxxxxxxxx>: > Hi, > > > On 2018-03-01 19:20:04 -0300, Andre Oliveira Freitas wrote: >> I was able to capture the backtrace again, now with libpq debugging symbols. >> >> Thread 15 (Thread 0x7f8cec068700 (LWP 68)): >> #0 0x00007f8d252a1d9b in __libc_recv (fd=150, buf=0x7f8cf0034410, >> n=16384, flags=623517083, flags@entry=0) at >> ../sysdeps/unix/sysv/linux/x86_64/recv.c:33 >> #1 0x00007f8d26689783 in recv (__flags=0, __n=<optimized out>, >> __buf=<optimized out>, __fd=<optimized out>) at >> /usr/include/x86_64-linux-gnu/bits/socket2.h:44 >> #2 pqsecure_raw_read (conn=conn@entry=0x7f8cf001e390, ptr=<optimized >> out>, len=<optimized out>) at >> /build/postgresql-10-rIfDLC/postgresql-10-10.2/build/../src/interfaces/libpq/fe-secure.c:230 >> #3 0x00007f8d26689863 in pqsecure_read >> (conn=conn@entry=0x7f8cf001e390, ptr=<optimized out>, len=<optimized >> out>) at /build/postgresql-10-rIfDLC/postgresql-10-10.2/build/../src/interfaces/libpq/fe-secure.c:217 >> #4 0x00007f8d266810ea in pqReadData (conn=conn@entry=0x7f8cf001e390) >> at /build/postgresql-10-rIfDLC/postgresql-10-10.2/build/../src/interfaces/libpq/fe-misc.c:683 >> #5 0x00007f8d2667e6f2 in PQconsumeInput (conn=0x7f8cf001e390) at >> /build/postgresql-10-rIfDLC/postgresql-10-10.2/build/../src/interfaces/libpq/fe-exec.c:1704 >> >> In this case, I also checked the pg_stat_activity and this particular >> connection on pg server side was idle for 15 minutes. I killed it >> using pg_terminate_backend, and then somehow the application resumed >> execution. > > There's something decidedly weird going on. Libpq always keeps the > connection in nonblocking mode internally. Blocking is implemented by > using select on the socket. So this should never block. > > Is there any chance parts of your application changes the sockets > block-y-ness? Is see your code is passing the socket around, so perhaps > that's happening somewhere outside of the file? > > >> As you can see, recv has received a non-zero value in flags > > Well, no, not really. recv() has a 0 flags, it's just libc's internal > implementation that appears to be showing up weird afaict. > > > Greetings, > > Andres Freund -- André Luis O. Freitas System Architect Rua do Rócio, 220 - Cj. 72 São Paulo - SP - 04552-000 55 11 4063 4222 afreitas@xxxxxxxxxxxxx www.callix.com.br
