Filipe, Thanks for point out. We will remove the option from io-threads.vol. -- gowda On Fri, Jan 16, 2009 at 7:03 PM, Filipe Maia <filipe at xray.bmc.uu.se> wrote: > Then I think it should be removed from the doc/examples/io-threads.vol > example, as it can be confusing. > That link with the options is very usefull, thank you. > > On Fri, Jan 16, 2009 at 14:00, Harshavardhana Ranganath > <harsha at zresearch.com> wrote: > > Hi Filipe Maia, > > > > io-threads translator doesn't have option "cache-size" . cache-size > > option should be used for io-cache translator. Read the following link to > > understand more about currently available options about each translator > > http://gluster.org/docs/index.php/Translators_options > > > > and following log > > "2009-01-14 12:22:43 W [write-behind.c:1363:init] brick: aggregate-size > > is not zero, disabling flush-behind" > > > > Is harmless. under current code base its loglevel is changed to DEBUG. > > > > On Wed, Jan 14, 2009 at 6:01 PM, Filipe Maia <filipe at xray.bmc.uu.se> > wrote: > >> > >> Hi, > >> > >> I've tried to use the doc/examples/write-behind.vol and io-threads.vol > >> examples in my unify configuration. > >> Here's the glusterfs-server.vol: > >> > >> > >> volume disk > >> type storage/posix # POSIX FS translator > >> option directory /export/data # Export this directory > >> end-volume > >> > >> volume ns > >> type storage/posix > >> option directory /export/ns > >> end-volume > >> > >> ### 'IO-threads' translator gives a threading behaviour to File I/O > >> calls. > >> # All other normal fops are having default behaviour. Loading this on > >> server side helps > >> # to reduce the contension of network. (Which is assumed as a GlusterFS > >> hang). > >> # One can load it in client side to reduce the latency involved in case > of > >> a > >> # slow network, when loaded below write-behind. > >> volume iot > >> type performance/io-threads > >> subvolumes disk > >> option thread-count 4 # default value is 1 > >> option cache-size 16MB # default is 64MB (This is per thread, so > >> configure it > >> # according to your RAM size and thread-count. > >> end-volume > >> > >> > >> ### 'Write-behind' translator is a performance booster for write > >> operation. Best > >> # used on client side, as its main intension is to reduce the network > >> latency > >> # caused for each write operation. > >> > >> volume brick > >> type performance/write-behind > >> subvolumes iot > >> option flush-behind on # default value is 'off' > >> option window-size 2MB > >> option aggregate-size 1MB # default value is 0 > >> end-volume > >> > >> # Volume name is server > >> volume server > >> type protocol/server > >> option transport-type tcp > >> option auth.addr.brick.allow * > >> option auth.addr.ns.allow * > >> subvolumes brick ns > >> end-volume > >> > >> > >> > >> I'm getting the following warnings with glusterfs-1.4.0rc7: > >> > >> tintoretto:~# tail /var/log/glusterfsd.log > >> 2009-01-14 12:22:43 W [write-behind.c:1363:init] brick: aggregate-size > >> is not zero, disabling flush-behind > >> 2009-01-14 12:22:43 W [glusterfsd.c:416:_log_if_option_is_invalid] > >> iot: option 'cache-size' is not recognized > >> tintoretto:~# > >> > >> What am I doing wrong? > >> > >> Filipe > >> > >> _______________________________________________ > >> Gluster-users mailing list > >> Gluster-users at gluster.org > >> http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users > > > > > > > > -- > > Harshavardhana > > [y4m4 on #gluster at irc.freenode.net] > > "Samudaya TantraShilpi" > > Z Research Inc - http://www.zresearch.com > > > > > > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://zresearch.com/pipermail/gluster-users/attachments/20090117/8a2eaf72/attachment.htm
