Hi Murali, In at least one case, I can provoke the error messages on the client. It was when I interrupt a simple compile "mpif90 hello.f90" on a pvfs2 mounted volume with ^C (because it took so long) ... and it is reproducible!
Best regards, -Minh. Murali Vilayannur wrote: > Hi Trach, > Hmm.. are you experiencing any I/O failures as a result of these? > It looks like some I/O timeouts may have caused the cancellations in > the flow layer or something.. I think Sam or Phil may have something > more insightful to say than me.. > thanks! > Murali > > On 1/25/07, Trach-Minh Tran <[EMAIL PROTECTED]> wrote: >> Dear all, >> >> I have upgraded to pvfs-2.6.1 about 1 month ago and all seems to work >> fine since then. However, from time to time I get the following messagea >> on /tmp/pvfs2-client.log: >> >> === >> [E 14:49:21.789014] fp_multiqueue_cancel: flow proto cancel called on >> 0xba49a8 >> [E 14:49:21.801303] handle_io_error: flow proto error cleanup started >> on 0xba49a8, error_code: -1610612737 >> [E 14:49:21.801381] handle_io_error: flow proto 0xba49a8 canceled 1 >> operations, will clean up. >> [E 14:49:21.801404] fp_multiqueue_cancel: flow proto cancel called on >> 0xba5110 >> [E 14:49:21.801422] handle_io_error: flow proto error cleanup started >> on 0xba5110, error_code: -1610612737 >> [E 14:49:21.801458] handle_io_error: flow proto 0xba5110 canceled 1 >> operations, will clean up. >> [E 14:49:21.801478] fp_multiqueue_cancel: flow proto cancel called on >> 0xba5fe0 >> [E 14:49:21.801495] handle_io_error: flow proto error cleanup started >> on 0xba5fe0, error_code: -1610612737 >> [E 14:49:21.801529] handle_io_error: flow proto 0xba5fe0 canceled 1 >> operations, will clean up. >> [E 14:49:21.801554] fp_multiqueue_cancel: flow proto cancel called on >> 0xba6748 >> [E 14:49:21.801570] handle_io_error: flow proto error cleanup started >> on 0xba6748, error_code: -1610612737 >> [E 14:49:21.801631] handle_io_error: flow proto 0xba6748 canceled 1 >> operations, will clean up. >> [E 14:49:21.801650] fp_multiqueue_cancel: flow proto cancel called on >> 0xba6eb0 >> [E 14:49:21.801670] handle_io_error: flow proto error cleanup started >> on 0xba6eb0, error_code: -1610612737 >> [E 14:49:21.801739] handle_io_error: flow proto 0xba6eb0 canceled 1 >> operations, will clean up. >> [E 14:49:21.801827] handle_io_error: flow proto 0xba49a8 error cleanup >> finished, error_code: -1610612737 >> [E 14:49:21.801852] handle_io_error: flow proto 0xba5110 error cleanup >> finished, error_code: -1610612737 >> [E 14:49:21.801873] handle_io_error: flow proto 0xba5fe0 error cleanup >> finished, error_code: -1610612737 >> [E 14:49:21.801893] handle_io_error: flow proto 0xba6748 error cleanup >> finished, error_code: -1610612737 >> [E 14:49:21.801916] handle_io_error: flow proto 0xba6eb0 error cleanup >> finished, error_code: -1610612737 >> === >> >> and at the same time on 5 of my data servers the following messages >> (produced here >> only on 1 server) appear in /tmp/pvfs2-server.log: >> >> === >> [E 14:49:21.804608] handle_io_error: flow proto error cleanup started >> on 0x119b830, error_code: -104 >> [E 14:49:21.812521] handle_io_error: flow proto 0x119b830 canceled 1 >> operations, will clean up. >> [E 14:49:21.812543] handle_io_error: flow proto 0x119b830 error >> cleanup finished, error_code: -104 >> === >> >> Do I have to worry about that? >> >> Thanks for any infos, -Minh. >> _______________________________________________ >> Pvfs2-users mailing list >> [email protected] >> http://www.beowulf-underground.org/mailman/listinfo/pvfs2-users >> > -- Dr. Trach-Minh Tran CRPP/SB/EPFL http://crppwww.epfl.ch Swiss Federal Institute of Technology CH-1015 Lausanne Office: http://plan.epfl.ch/?room=PPB111 Tel: +41 21 693 4569 Fax: +41 21 693 5176 _______________________________________________ Pvfs2-users mailing list [email protected] http://www.beowulf-underground.org/mailman/listinfo/pvfs2-users
