Xsanity Forums

Troubleshooting :: Trouble adding back clients and BMDC

JonBerry — Fri, 03 Feb 2012 18:55:57 GMT

Author: JonBerry
Subject: Trouble adding back clients and BMDC
Posted: Fri Feb 03, 2012 1:55 pm (GMT -5)

Came back from vacation to find my BMDC throwing the error of fsmpm not running. Shut down all the systems, tried to remove and re-add the BMDC to the XSAN, same error, although I could add it, but only as a client. Still same error of fsmpm not running.

Started adding back clients to the XSAN. Couple worked fine no problem, but the rest won't add back because of fsmpm not running. It isn't running in activity monitor, but when I try and add the computer back to the SAN, it pops up as running, until XSAN Admin gives up trying to add the computer.

I've re-installed, repaired disk permissions, tried different versions of XSAN. Nothing works. Thoughts?

Running 10.6.4 on clients, 10.6.2 on BMDC, 10.6.4 Server on MDC. XSAN version 2.2.1.

Troubleshooting :: RE: Shared Computers Problem - why?

nrausch — Fri, 03 Feb 2012 17:56:17 GMT

Author: nrausch
Posted: Fri Feb 03, 2012 12:56 pm (GMT -5)

Usually that indicates that this is the 2nd, or 3rd, or 4th, etc...
instance of that computer seen on the network. If these are XSAN clients, they will have 2 ethernet interfaces yes?
Metadata and Public Internet.

Be SURE that if the machines have airport cards, that the airport interface is completely disabled.
If you're using a VLAN for your 2 wired networks, be sure both subnet masks in your system prefs are set to 255.255.255.0

I find this is generally caused by one machine having 2 interfaces on the same network.

The other possible cause is if the systems were imaged or cloned, and the hostname file wasn't properly changed.

Usually it is the former though...

Software :: RE: MDC / XSAN update methodology

singlemalt — Fri, 03 Feb 2012 14:20:46 GMT

Author: singlemalt
Posted: Fri Feb 03, 2012 9:20 am (GMT -5)

Hi Elliot,
I see nothing wrong with your proposed method of updating. Only thing I would add
is to make a backup copy of the /Library/Filesystems/Xsan/config folder on one of the
MDCs and a copy of sudo cvlabel -c> ~/Desktop/lunlabels.txt somewhere. If you have a
copy of the <vol_name>.cfg files, config.plist (which has all your xsan serial numbers
see http://support.apple.com/kb/HT4391) and a cvlabels file you can pretty much resurrect
any xsan volume. If you have those you can even uninstall xsan on everything and still
bring it all back, just as long as somebody doesn't do something stupid like reformat
any luns.

Software :: MDC / XSAN update methodology

elliott — Fri, 03 Feb 2012 13:57:15 GMT

Author: elliott
Subject: MDC / XSAN update methodology
Posted: Fri Feb 03, 2012 8:57 am (GMT -5)

New user here!

Recently we've been having some issues with finder crashing when copying files (and also via terminal with both cp and cvcp), and in general our xsan environment has been a little unreliable, ghost volumes, spotlight crashes, random failovers etc.

We are running 10.6.4 and xsan 2.2.1 across all clients and mdc's, so I thought now might be the time to update everything to 10.6.8v1.1 and xsan 2.2.2 (there are added benefits too, the app store on clients and the latest nvidia drivers for our smacs).

I'm also hoping to do a clean reinstall of the mdc's as they been through a lot recently (we have added more storage blocks and built a new san and mirror to replace the old ones and have been renaming volumes, luns etc).

My question is: what would be the correct methodology for flattening and updating the mdc's and then clients?

Initially, I'm thinking of the below:

Unmount all clients and remove as computers from xsan, leaving just the mdc's running and connected,
Failover both volumes to mdc 1,
Remove mdc 2 as a controller and computer,
Disconnect ethernet and fibre and reinstall mdc 2 and update to 10.6.8v1.1, install xsan 2.2.2,
Add as a controller and failover both volumes to mdc 2,
Remove mdc1 as a controller and reinstall and update os and xsan,
Add as a controller,
Then finally, update all clients to 10.6.8v1.1 and xsan 2.2.2 and add them as clients,

The only thing that i'm concerned about here is having (if only for a short time) one mdc at 10.6.4 / 2.2.1 and the other on 10.6.8 / 2.2.2.

I will have a 3rd unused xserve that I could potentially use to host both volumes whilst I'm updating them.

I realise that simply updating the os without flattening them would be much easier, but we would rather clean install them.

Our setup consists of the following:

2 x MDC's - Xsevres running 10.6.4 and Xsan 2.2.1
39 x Clients running 10.6.4 and 2.2.1
192TB active storage SAN
252TB active storage Mirror

Any suggestions would be very much appreciated!

Thanks

Troubleshooting :: Shared Computers Problem - why?

giomla — Fri, 03 Feb 2012 13:31:44 GMT

Author: giomla
Subject: Shared Computers Problem - why?
Posted: Fri Feb 03, 2012 8:31 am (GMT -5)

I have this odd "shared" computers behavior within Finder and Xsan Admin. Every now and then computer name gets higher and higher number in the brackets.

I cannot figure out why this is happening. Even when i remove the number in the system preferences -> shared, on the local machine, number appears after a while again. Could it be some IP or afp issue? Don't know even in what direction to think...

Thanks a lot for your help.

Hardware :: link 2 san with fiber

loccoliv — Thu, 02 Feb 2012 12:48:38 GMT

Author: loccoliv
Subject: link 2 san with fiber
Posted: Thu Feb 02, 2012 7:48 am (GMT -5)

hi,

i've got 2 san, our good old 1.4.2 with 3 volumes and a 2.2.2, both completely independent.

I have to copy 20tb from one volume on the 2.2.2 san to a 1.4.2 hosted volume, i do this thru ethernet, it takes hours

Is there a simple way to interconnect the 2 qlogic switches with a coulpe of fiber cables?

thnx

Troubleshooting :: RE: XSan 2.3 Disk Image issues

JSal — Thu, 02 Feb 2012 01:18:42 GMT

Author: JSal
Posted: Wed Feb 01, 2012 8:18 pm (GMT -5)

Yes! I just came on-line to post the same. 10.7.3 fixes this issue. Thank you singlemalt for confirming the bug.

Troubleshooting :: RE: DNS Question regarding Metadata network

MattRK — Thu, 02 Feb 2012 01:17:54 GMT

Author: MattRK
Posted: Wed Feb 01, 2012 8:17 pm (GMT -5)

Thanks! I'll give that a try.

Promise VTrak :: RE: Vtrak x30s

abstractrude — Thu, 02 Feb 2012 00:40:30 GMT

Author: abstractrude
Posted: Wed Feb 01, 2012 7:40 pm (GMT -5)

10.7.2. its a test setup. hosts virtual machines and does backup and replication.

Vendor: LSILogic
Product: LSI7204EP
Revision: Firmware 1.3.20.0, EFI 1.05.05.00
Bus: PCI
Slot: Slot-2
Initiator Identifier: 126

Manufacturer: Promise
Model: VTrak E830f
Revision: 402
ALUA Support: Unsupported
Logical Unit Access State: Active Optimized
Load Balancing Algorithm: Least Bytes

Troubleshooting :: RE: XSan 2.3 Disk Image issues

singlemalt — Wed, 01 Feb 2012 21:43:46 GMT

Author: singlemalt
Posted: Wed Feb 01, 2012 4:43 pm (GMT -5)

I was able to reproduce this earlier but it looks like the 10.7.3 update resolved
it. The update just went live a little while ago.

Promise VTrak :: RE: Vtrak x30s

flumignan — Wed, 01 Feb 2012 18:22:48 GMT

Author: flumignan
Posted: Wed Feb 01, 2012 1:22 pm (GMT -5)

abstractrude wrote:

I have one of the non apple boxes attached to an xserve. it works just fine.

Thanks for sharing your experience. Could you tell me a bit about your setup? Are you running 10.6 or 10.7 Server? Are you connecting directly to the Xserve? Ours is an Xserve (Early 2009). What about your fibre card? Ours is the OEM fibre card from Apple, which System Profiler says is an LSILogic LSI7404EP with firmware 1.3.23.0 and EFI 1.05.06.00. According to Promise's compatibility matrix, they cite firmware 1.03.27 as being certified, but I worry that the older firmware might cause stability problems.

I appreciate the info very much -- it's helpful.

Promise VTrak :: RE: Need 32 new 2TB hard drives for a E610 and a J610

dickot — Wed, 01 Feb 2012 15:16:15 GMT

Author: dickot
Posted: Wed Feb 01, 2012 10:16 am (GMT -5)

Hi - just wanted to mention that it is not a "failure" to release new firmware for a product. The E610 for Mac has been very stable and no fixes are required, hence no release since September 2009. When either Apple or Promise have a reason to do it, there will be a service release.
As an "Apple" SKU the qualification of drives for the product is in their hands, however there is no reason you can't peruse the non-Apple compatibility list to see what drives Promise has qualified for that SKU.
As you have noted, any other drives wont be covered by AppleCare or Promise; you will have to take care of RMAs yourself...

Promise VTrak :: RE: Vtrak x30s

abstractrude — Wed, 01 Feb 2012 02:45:56 GMT

Author: abstractrude
Posted: Tue Jan 31, 2012 9:45 pm (GMT -5)

I have one of the non apple boxes attached to an xserve. it works just fine. my understanding is that the main difference is there is more RAM in the apple models.

Promise VTrak :: RE: Vtrak x30s

flumignan — Wed, 01 Feb 2012 00:31:11 GMT

Author: flumignan
Posted: Tue Jan 31, 2012 7:31 pm (GMT -5)

Has anyone deployed the non-Apple spec x30 on an Xserve? I know they're not supported, and I know the firmware is different.

We connect directly to our 2010 Xserve via fibre channel. We don't use Xsan, we don't do video. Just lots and lots of small medical-related files that get manipulated by two dozen researchers via NFS, AFP and SMB.

I'm working with a research group that needs a lot of disk space, but they have a small budget. We were thinking about the VTrak 830, but even with the .edu discount, the "Made for Mac" version is $23,500 and must come with 24 x 2TB drives. We could pick up the single-controller VTrak 830fS for $7,700 and 24 3TB Hitachi SATA drives for $8,760 and pay $16,450. That's 30% less for 33% more space, minus one controller. We can even throw in an ATTO 8Gbps fibre card and be ahead. Plus all our drives will have 5-year warranties.

We already have nearly 100TB of storage mixed between non-Apple VTrak 610 chassis and "Made for Mac" ones, too. In our experience, they're indistinguishable. The x30 compatibility matrix says the HBA in the Xserve is not quite supported (they support the same make/model, but with newer firmware on the fibre card). I'm just wondering if it works. Obviously, the most terabytes per dollar is our primary consideration, above performance and supportability.

Promise VTrak :: RE: Need 32 new 2TB hard drives for a E610 and a J610

flumignan — Wed, 01 Feb 2012 00:29:02 GMT

Author: flumignan
Posted: Tue Jan 31, 2012 7:29 pm (GMT -5)

We've installed Western Digital RE4, 2 TB (WD2002FYPS) in our Apple-spec x10 E and J-class chassis without incident. I can't exactly remember when we installed them -- maybe a year and a half ago? We've replaced maybe two drives since then, out of 32, I think. They have a five year warranty and replacements are super easy.

Troubleshooting :: RE: Finder crashes after xsan finder copy, Apple is stumped

lucasnap — Tue, 31 Jan 2012 14:27:13 GMT

Author: lucasnap
Posted: Tue Jan 31, 2012 9:27 am (GMT -5)

Hi Starbytes,

need some help?

Lucas

Troubleshooting :: RE: DNS Question regarding Metadata network

ACSA — Tue, 31 Jan 2012 14:17:31 GMT

Author: ACSA
Subject: Re: DNS Question regarding Metadata network
Posted: Tue Jan 31, 2012 9:17 am (GMT -5)

MattRK wrote:

I have a quick question. I seem to be having some issues with DNS. As it stands i'm having trouble getting new clients added to the san and i think DNS is to blame. I've read that we should setup DNS A records for both nics. (Corporate network & Metadata network) My question is this:

Do i need to setup these A records for the two different NICs in the same DNS zone? (In our existing corporate DNS zone for example) Or should I create a separate DNS zone just for the metadata network records? (E.g. Corporate network DNS zone = corporate.net & Metadata DNS zone = metadata.corporate.net)

If not and I need to create both of these A records in the existing corporate zone, do i need to give each NIC's A record a different name or keep it the same? (E.g. edit1=10.0.0.100 & edit1=192.168.0.100 OR edit1=10.0.0.100 & edit1-meta=192.168.0.100)

I just want to make sure i'm setting DNS up correctly.

Thanks!

You will have to create 2 DNS zones, one for corporate and one for the metadata:

i.e.: corperate.net and corperate.meta.

Create for each IP address an A record in both zones.

Each zone will reflect to a different IP subnet:

corperate.net = 192.168.x.x
corperate.met = 10.x.x.x

Whatever subnet works for you is fine as long as they are different.

Troubleshooting :: RE: Finder crashes after xsan finder copy, Apple is stumped

tomwelch — Tue, 31 Jan 2012 11:12:02 GMT

Author: tomwelch
Posted: Tue Jan 31, 2012 6:12 am (GMT -5)

Hi Starbytes

We upgraded to 2.2.2 back in September 2011, and we haven't experienced the problems we were having almost daily since. Fingers crossed, that's it.

Hope this helps

Tom

Troubleshooting :: RE: Finder crashes after xsan finder copy, Apple is stumped

Starbytes — Tue, 31 Jan 2012 11:07:54 GMT

Author: Starbytes
Posted: Tue Jan 31, 2012 6:07 am (GMT -5)

Any news on the subject? We are having the same problems with a customer..

Promise VTrak :: RE: Share Storage for Ingest System

nrausch — Mon, 30 Jan 2012 22:13:04 GMT

Author: nrausch
Posted: Mon Jan 30, 2012 5:13 pm (GMT -5)

Absolutely!

Sounds like you have everything you need except the MDC servers and a dedicated Ethernet network for Metadata.

Attach the 4 ingest servers and the edit machine to same same storage, and if your software supports chunking, you could even edit while you're still capturing.

Good Luck!

Troubleshooting :: RE: Coordinator List Mismatch

singlemalt — Mon, 30 Jan 2012 22:09:56 GMT

Author: singlemalt
Posted: Mon Jan 30, 2012 5:09 pm (GMT -5)

I believe the fsnameserver list would get randomly changed anytime a client
was added to the SAN. In Lion they will be in numerically descending. I don't it matters which order they as in as long as they are in the same order.

Promise VTrak :: RE: Share Storage for Ingest System

poucedeleon — Mon, 30 Jan 2012 22:00:19 GMT

Author: poucedeleon
Posted: Mon Jan 30, 2012 5:00 pm (GMT -5)

The streams will be edited after they are captured. My edit system have a Direct Attached Raid w/FC where I will be doing the editing. It sounds like it would be in my best interest to have a XSAN installed.

Troubleshooting :: RE: Coordinator List Mismatch

rberd — Mon, 30 Jan 2012 20:49:08 GMT

Author: rberd
Posted: Mon Jan 30, 2012 3:49 pm (GMT -5)

Thanks for your response.

Yes, it does look like some machines have a different order than others. Assuming the Primary MDC IP should be listed first (at the top of the list) and Secondary MDC IP second? If that is the case than they are wrong on the controllers as well.

Typically, what causes this kind mismatch to happen?

Thank you!

Troubleshooting :: RE: XSan 2.3 Disk Image issues

JSal — Mon, 30 Jan 2012 20:45:17 GMT

Author: JSal
Posted: Mon Jan 30, 2012 3:45 pm (GMT -5)

Sorry the error is "no mountable file system".

Additionally, if I mount the same XSan volume through an afp reshare and then attempt to mount the disk image all works normally. This problem seems to exist just on direct Xsan access and just through finder.

Any ideas would be helpful.

Troubleshooting :: XSan 2.3 Disk Image issues

JSal — Mon, 30 Jan 2012 20:25:34 GMT

Author: JSal
Subject: XSan 2.3 Disk Image issues
Posted: Mon Jan 30, 2012 3:25 pm (GMT -5)

Is anyone else seeing this? Running XSan 2.3 controllers and clients. I am copying installer disc images to the san, as well as attempting to use disk images as Media volumes for AVID clients. The first time I mount the image al seems to work well. But after that I am unable to remount the image. Keep getting a "filesystem not available" error. If I copy the disk image off the drive I am able to mount it again and all data seems to be intact. Same behavior in dmg, sparsebundle and iso images.