On 06/17/2015 09:28 PM, Ravishankar N wrote:> Hello, > > I sent a patch [1] to print the backtrace generated by > 'gf_log_callingfn' (only) in a human-readable way. While it improves > readability by printing every function of the backtrace in a separate > line (which is how a backtrace *should* look), it violates the fact > that every log in gluster gets logged in a single line. Does anyone > see any problem with this? Especially sys-admins who might have > scripts/ mechanisms to parse logs? Note that the patch affects only > messages logged using gf_log_callingfn() and is avoids eyesore.Grepping logs will be difficult with this change. May be we can use something like: sed 's/(-->/\n(-->/g' <logfile.log> to achieve the same? Pranith> > Thanks, > Ravi > > > [1] http://review.gluster.org/#/c/11217/1 > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://www.gluster.org/mailman/listinfo/gluster-users
Sent from one plus one On Jun 17, 2015 10:10 PM, "Pranith Kumar Karampuri" <pkarampu at redhat.com> wrote:> > > > On 06/17/2015 09:28 PM, Ravishankar N wrote: >> >> Hello, >> >> I sent a patch [1] to print the backtrace generated by'gf_log_callingfn' (only) in a human-readable way. While it improves readability by printing every function of the backtrace in a separate line (which is how a backtrace *should* look), it violates the fact that every log in gluster gets logged in a single line. Does anyone see any problem with this? Especially sys-admins who might have scripts/ mechanisms to parse logs? Note that the patch affects only messages logged using gf_log_callingfn() and is avoids eyesore.> > Grepping logs will be difficult with this change. May be we can usesomething like:> sed 's/(-->/\n(-->/g' <logfile.log> to achieve the same?It doesn't bother me too much if some changes are required in applying filters, you will need this change if you try to improve readability. Being a developer sometimes I get confused with this log, at least its not readable. I would like to see this change in (I am yet to review it though).> > Pranith > >> >> Thanks, >> Ravi >> >> >> [1] http://review.gluster.org/#/c/11217/1 >> _______________________________________________ >> Gluster-users mailing list >> Gluster-users at gluster.org >> http://www.gluster.org/mailman/listinfo/gluster-users > > > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://www.gluster.org/mailman/listinfo/gluster-users-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20150617/85ccb1a6/attachment.html>
On 06/17/2015 10:10 PM, Pranith Kumar Karampuri wrote:> > > On 06/17/2015 09:28 PM, Ravishankar N wrote: >> Hello, >> >> I sent a patch [1] to print the backtrace generated by >> 'gf_log_callingfn' (only) in a human-readable way. While it improves >> readability by printing every function of the backtrace in a separate >> line (which is how a backtrace *should* look), it violates the fact >> that every log in gluster gets logged in a single line. Does anyone >> see any problem with this? Especially sys-admins who might have >> scripts/ mechanisms to parse logs? Note that the patch affects only >> messages logged using gf_log_callingfn() and is avoids eyesore. > Grepping logs will be difficult with this change. May be we can use > something like:Could you elaborate? IMO, `grep -A10 gf_log_callingfn <logfile.log>` to list the backtrace seems a lot easier than typing that substitution pattern.> sed 's/(-->/\n(-->/g' <logfile.log> to achieve the same?> > Pranith >> >> Thanks, >> Ravi >> >> >> [1] http://review.gluster.org/#/c/11217/1 >> _______________________________________________ >> Gluster-users mailing list >> Gluster-users at gluster.org >> http://www.gluster.org/mailman/listinfo/gluster-users >