On Fri, 2007-10-12 at 13:59 -0700, Greg Kurtzer wrote: > I should have also mentioned that specifying the port works just fine: > > # perfquery 1 1 > # Port counters: Lid 1 port 1 > PortSelect:......................1 > CounterSelect:...................0x0000 > SymbolErrors:....................0 > LinkRecovers:....................0 > LinkDowned:......................0 > RcvErrors:.......................0 > RcvRemotePhysErrors:.............0 > RcvSwRelayErrors:................0 > XmtDiscards:.....................0 > XmtConstraintErrors:.............0 > RcvConstraintErrors:.............0 > LinkIntegrityErrors:.............0 > ExcBufOverrunErrors:.............0 > VL15Dropped:.....................0 > XmtData:.........................2543853998 > RcvData:.........................2544066226 > XmtPkts:.........................9673830 > RcvPkts:.........................9673818 > # perfquery 2 1 > # Port counters: Lid 2 port 1 > PortSelect:......................1 > CounterSelect:...................0x0000 > SymbolErrors:....................0 > LinkRecovers:....................0 > LinkDowned:......................0 > RcvErrors:.......................0 > RcvRemotePhysErrors:.............0 > RcvSwRelayErrors:................0 > XmtDiscards:.....................0 > XmtConstraintErrors:.............0 > RcvConstraintErrors:.............0 > LinkIntegrityErrors:.............0 > ExcBufOverrunErrors:.............0 > VL15Dropped:.....................0 > XmtData:.........................2783649428 > RcvData:.........................2762924012 > XmtPkts:.........................30190437 > RcvPkts:.........................30190457
Are these one port HCAs ? Wonder what PerfMgt ClassPortInfo:CapabilityMask is. Could I get you to add some "debug" code to display this ? perfquery should probably check this (when -a is being used). The scripts need more changes to work in this mode as I think they assume "all ports" is supported. -- Hal > > > On Oct 12, 2007, at 1:58 PM, Greg Kurtzer wrote: > > > > > # perfquery 1 -a > > perfquery: iberror: failed: perfquery > > # perfquery 2 -a > > perfquery: iberror: failed: perfquery > > > > Thanks! > > > > > > On Oct 12, 2007, at 1:56 PM, Hal Rosenstock wrote: > > > >> Greg, > >> > >> On Fri, 2007-10-12 at 13:27 -0700, Greg Kurtzer wrote: > >>> I am getting this error with the current packaged releases: > >>> > >>> > >>> # ibcheckerrors > >>> perfquery: iberror: failed: perfquery > >>> Error check on lid 2 (Topspin DDR-HCAe LX x8) port all: FAILED > >>> perfquery: iberror: failed: perfquery > >>> Error check on lid 1 (ibtest1 HCA-1) port all: FAILED > >> > >> What do: > >> perfquery 2 -a > >> and > >> perfquery 1 -a > >> show ? > >> > >> -- Hal > >> > >>> > >>> ## Summary: 2 nodes checked, 0 bad nodes found > >>> ## 2 ports checked, 0 ports have errors beyond threshold > >>> > >>> > >>> ibcheckerrors seems to be calling the script ibcheckerrs with a port > >>> of 255 and ibcheckerrs is calling the binary perfquery with the "-a" > >>> option (which is apparently not working as it is supposed to query > >>> all ports) which then calls into libibmad (port_performance_query > >>> and > >>> then pma_query). > >>> > >>> I am using libibmad-1.1.2 and infiniband-diags-1.3.2 which are the > >>> latest versions from the download directory on the web site. Is it > >>> possible this is a bug, or did I grab incompatible versions of the > >>> packages? > >>> > >>> Many thanks! > >>> > >>> Greg > >>> > >>> note: Please CC me with any replies as I am not (yet) a list member. > >>> Thanks again. :) > >>> > >>> > >>> -- > >>> Greg Kurtzer > >>> [EMAIL PROTECTED] > >>> > >>> > >>> _______________________________________________ > >>> general mailing list > >>> [email protected] > >>> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general > >>> > >>> To unsubscribe, please visit http://openib.org/mailman/listinfo/ > >>> openib-general > > > > -- > > Greg Kurtzer > > [EMAIL PROTECTED] > > > > > > > > -- > Greg Kurtzer > [EMAIL PROTECTED] > > > _______________________________________________ > general mailing list > [email protected] > http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general > > To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
