Hello all, For those of you that have been following this thread.
Since my last post I have been in contact with both Overland and SUN in regards to this performance problem that i have. Latest news is that SUN have contacted Overland to borrow an Overland REO 9000 unit for making some inhouse testing. I am still in contact with SUN who has made a Management Escalation so if this means that the problem has left France is uncertain but my guess is YES. The support in general here in France for the iscsi seems to be very limited. As to our supplier (in Paris) of the Overland unit goes they have proposed to give us a different iSCSI unit to use with our Solaris systems and especially 10. They can not understand why there is a performance problem between Solaris 10 native iscsi driver and their Overland REO 9000. Is is after all working perfect under Solaris 8 with Cisco drivers, HP/UX and Windows. Due to their limited knowledge in Solaris/Unix they will not pursue this matter any longer it seems. Besides, we are the only customer running Solaris 10 with their HW. I am continuing with tests between Windows and the iSCSI. David Weibel who has been helping me a lot in his sparetime has proposed a few new things that i will try as soon as i have some time to play with a Windows server. We are going to collect 3 very short traces. a. The first would be to record the login sequence between the windows initiator and target. b. The second would be a couple seconds after the initiator is connected and the system is idle. c. The third would be a very short trace during IO. For those of you who are still interested i will post any news that may arrive from SUN if and when they have some test results to provide. Thank you all of you that have been giving advice and helping me out during this time. Regards, Pierre This message posted from opensolaris.org _______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
