Hi Nils, Attached is the Openoffice source. I think we should measure some failover and switchover times with iSCSI devices in Sun Cluster in different configurations and see when it becomes unacceptable for cluster.
A question on the tunable timeouts that you proposed to the storage community. Did the changes in the iSCSI initiator for IDM in COMSTAR help at all ? In other words, are there tunables in IDM that can be used instead ? thanks, Augustus. Nils Goroll wrote: > Hi Augustus, > >> Thanks for taking the time to restructure the document. The new >> organization looks good to me. > > I am glad to hear that you agree. If you could send me the OpenOffice source, > I'd be willing to put these changes into the right format. Also, Steve has > asked > me in private not to change the requirements numbering, so that would need to > be > corrected. > >> You promised to get some data that we can use for this, right ? > > You are probably referring to this message from Oct 4th > > (quoting myself) > >> Would you be willing to help in measuring some of these for Colorado ? > > > > I can do that. I might get access to a (real, not emulated) NetApp box > (though > > I > > will have to ask, cannot promise) plus I can do everything that is possible > > using an SNV_98 notebook plus an s10 workstation (w2100z). > > > > But at this point I am pretty sure that the currently hardcoded timeout > values > > will give us unacceptable results, so I would consider it much more > sensible > > to > > integrate at least some way of timeout tuning into the iSCSI stack and then > > conduct tests which will also give an idea about which timeout values > would be > > sensible. > > > > Please let me know what you think > > Could you please let me know what you think the tests should show? > > Nils > _______________________________________________ > ha-clusters-discuss mailing list > ha-clusters-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/ha-clusters-discuss -- Augustus Franklin, Solaris Cluster http://blogs.sun.com/augustus -------------- next part -------------- A non-text attachment was scrubbed... Name: requirements_ext_1.odt Type: application/vnd.oasis.opendocument.text Size: 21188 bytes Desc: not available URL: <http://mail.opensolaris.org/pipermail/ha-clusters-discuss/attachments/20081023/7a785417/attachment.odt>
