On 2 Apr 2009 at 6:34, Gonçalo Borges wrote: [...] > In conclusion, it seems that at a given time there are attempts to > access beyond end of device. I don't know who is the guilty guy, if > the kernel itself if the iscsi framework. Then the ISCSI controller > start to fail also with IO expired messages.
It might be good to know what scsiinfo (or similar) says about the size of the LUN at the start of yout tests. Likewise, show what "fdisk -l" tells about the partitions, and finally what "df -k" tells about the capacity of the file system. Then one could compare those sizes to those reported by the kernel. Maybe the setup just wrong, and it takes a while until the end of the device is reached. Then I would start slowly, i.e. with one izone running on one client. BTW, what do you want to measure: the kernel throughput, the network throughput, the iSCSI throughput, the controller throughput, or the disk throughput? You should have some concrete idea before starting the benchmark. Also with just 12 disks I see little sense in having that many threads accessign the disk. To shorten a lengthy test, it may be advisable to reduce the system memory (iozone recommands to create a file size at least three times the amount of RAM, end even 8GB on a local disk takes hours to perform) Regards, Ulrich > > Do you have suggestion of what can I be doing wrong? --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "open-iscsi" group. To post to this group, send email to open-iscsi@googlegroups.com To unsubscribe from this group, send email to open-iscsi+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/open-iscsi -~----------~----~----~----~------~----~------~--~---