This is just a quick reference to create a multiplathing iSCSI setup…
Create two virtual kernel switches, one called “iSCSI-1″ and the other called “iSCSI-2″ (and so on if you have more nics)
Then per kernel portgroup ensure that only one of the nics is active. For the “iSCSI-1″ portgroup configure it to override the virtual switch settings and move nic 0 to active and nic 1 to unused. For the “iSCSI-2″ portgroup configure it to override the virtual switch settings and move nic 1 to active and nic 0 to unused.
Now you have to run some esxcli commands to gel things together…. alt-f1 on the esxi console and type “unsupported” followed by your root password. The following is the command i have to run to get both my portgroups to work together on the iscsi hba…. (you can check your vmk number from the networking config screen)
esxcli swiscsi nic add -n vmk1 -d vmhba33
esxcli swiscsi nic add -n vmk2 -d vmhba33
For each target then change the path selection method to “round-robin”.
Go back to storage adapters and click “rescan”
If you want all future iSCSI targets to automatically use round-robin you must also run the following from commandline… (this is for our HP Lefthand, your “storage array type” may be different. Its listed under your target details). Basically sets round robin as a default for this type of array. In general you should do this first before presenting any LUNs etc, else you may have to bounce your box.
esxcli nmp satp setdefaultpsp –satp VMW_SATP_DEFAULT_AA –psp VMW_PSP_RR
I’ve got some old 250Gb drives that are starting to show their age. I’ve currently got them setup in a 3x RAID 0 config which presents about 750Gb of space.
I’ve got everything on a single partition (meh, i’m lazy). I’ve done various speed tests in the current setup (with all space allocated), but i thought i’d re-image onto a short stroke partition.
I only use about 150Gb of space on my main machine (most of my data is on another box), so i’m going to try creating a 200Gb partition to test if this provides any kind of performance boost.
So reducing my raid 0 from 750Gb to 214Gb, and here are the results…
Before with all 750Gb presented…
Same disks but short stroked to 214Gb….
Conclusion : Yip, seems like its worth it if you have the spare space. Average throughput is up by 10MB/s and seek has improved by almost a third loosing 4ms.
You will get even more of an improvement if you can use a smaller % of capacity per drive and / or more drives for your stripe.
Updated : 07/02/2010
btw – the above was without write-back cache enabled…. if i turned that on i got the following…
Want iSCSI in opensolaris?
Grab SUNWiscsitgt via package manager.
enable the server via svcadm;
svcadm enable iscsitgt
create your zfs iscsi pool; (this command will limit iscsi drive to 500GB in size)
zfs create -V 500G tank/iscsi
set isci on via zfs command;
zfs set shareiscsi=on tank/iscsi
check that target is up and running;
iscsitadm list target -v
Done. Should be able to connect via ip from another machine. I have not covered CHAP or any client side configuration. Assumed isolated LAN.
If you want to give zfs a go and also want a dedicated file server this is the solution.
Check it out here;
http://www.freenas.org/
I’m still using opensolaris though as i like running a few virtualbox machines on the same box.
Recently i hit what i thought was a huge disaster with my ZFS array. Essentially i was unable to import my zpool without causing the kernel to panic and reboot. Still unsure of the exact reason, but it didn’t seem to be due to a hardware fault. (zpool import showed all disks as ONLINE)
When i tried to import with zpool import -f tank the machine would lockup and reboot (panic).
The kernel panic; (key line)
> genunix: [ID 361072 kern.notice] zfs: freeing free segment (offset=3540185931776 size=22528)
Nothing i could do would fix it… tried both of these options in the system file with no success;
set zfs:zfs_recover=1
set aok=1
After a quick email from a Sun Engineer (kudos to Victor), the zdb command line that fixed it;
zdb -e -bcsvL <poolname>
zdb is a read only diagnostic tool, but seemed to read through the sectors that had the corrupt data and fix things?? (not sure how a read only tool does that) – the run took well over 15hrs.
Updated: 20/10/2009
Apparently if you have set zfs:zfs_recover=1 in your system file the zdb command will operate in a different manner fixing the issues it encounters.
Remember to run a zpool scrub <poolname> if you are lucky enough to get it back online.