Correct, there's clearly some other cases which sets ARC_L2COMPRESS on b_flags for write, but I cant see where unless its some how gets through from the read cases. Attached is a patch which removes all paths which set ARC_L2COMPRESS on b_flags for both read and write. Revert the last patch and try this one. To be clear l2_compress_failures are only a failure to compress the data, which is not an error. Have you seen any l2_io_error or l2_cksum_bad since applying the ashift patch? Regards Steve ----- Original Message ----- From: "Vitalij Satanivskij" <satan_at_ukr.net> > > So that's patch not disabling compresion on l2arc? > > > Steven Hartland wrote: > SH> I would have expected zero for all l2_compress values. > SH> > SH> Regards > SH> Steve > SH> > SH> ----- Original Message ----- > SH> From: "Vitalij Satanivskij" <satan_at_ukr.net> > SH> To: "Steven Hartland" <killing_at_multiplay.co.uk> > SH> Cc: "Vitalij Satanivskij" <satan_at_ukr.net>; "Dmitriy Makarov" <supportme_at_ukr.net>; <freebsd-current_at_freebsd.org> > SH> Sent: Friday, October 25, 2013 8:32 AM > SH> Subject: Re: ZFS L2ARC - incorrect size and abnormal system load on r255173 > SH> > SH> > SH> > Just after system reboot with this patch > SH> > > SH> > found > SH> > kstat.zfs.misc.arcstats.l2_compress_successes: 6083 > SH> > kstat.zfs.misc.arcstats.l2_compress_zeros: 1 > SH> > kstat.zfs.misc.arcstats.l2_compress_failures: 296 > SH> > > SH> > compression on test pool (where I'm test this patch) is lz4 > SH> > > SH> > so is it ok ? > SH> > > SH> > > SH> > > SH> > Steven Hartland wrote: > SH> > SH> If you are still seeing high load try commenting out the following > SH> > SH> which should disable l2arc compression. > SH> > SH> sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c > SH> > SH> if (l2arc_compress) > SH> > SH> hdr->b_flags |= ARC_L2COMPRESS; > SH> > SH> > SH> > > SH> > SH> ================================================ > SH> This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the > event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any > information contained in it. > SH> > SH> In the event of misdirection, illegible or incomplete transmission please telephone +44 845 868 1337 > SH> or return the E.mail to postmaster_at_multiplay.co.uk. > SH> > SH> _______________________________________________ > SH> freebsd-current_at_freebsd.org mailing list > SH> http://lists.freebsd.org/mailman/listinfo/freebsd-current > SH> To unsubscribe, send any mail to "freebsd-current-unsubscribe_at_freebsd.org" > ================================================ This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. In the event of misdirection, illegible or incomplete transmission please telephone +44 845 868 1337 or return the E.mail to postmaster_at_multiplay.co.uk.
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:43 UTC