Page 3 of 3 FirstFirst 1 2 3
Results 41 to 48 of 48

Thread: Mac Pro "Westmere" 3.33GHz Photoshop Workstation

  1. #41
    Join Date
    Aug 2002
    Posts
    1,010

    Default

    I am running ML. Someone has to do so to find out where the problems are -
    Yes, I can appreciate that.

    Usually, I wait until the OS version gets to a .5 or .6 before moving on up, but I had figured that Lion was sort of the 'beta' version of ML. In retrospect, it seemed like a hybrid- a bridge, with ML a completely different new universe..

    I feel really fortunate I can fall back. I have a lot of sympathy for those with newer machines who have no choice.. Watching one of the Logic Pro forums (which works great in SL but has a serious show-stopping flaw in Lion and ML), there are long-time users abandoning the app because Apple apparently cannot fix the problem without a major re-write of the app. It's sad to see. And those users cannot fall back to Snow Leopard..

    Oh well, eventually I'll have to move up myself but hopefully all the big kinks will be cured by then!

  2. #42
    Join Date
    Aug 2002
    Posts
    1,010

    Default A Little Tech Saga for the New Year...

    Hi everybody, it's been awhile..

    So I've had some interesting tech issues with my 6-core Mac Pro 3.33GHz lately..
    Here's a brief rundown:

    First, the SeriTek e6G 2-port SATA pci card failed, and in the process, appeared to take down my external 4-drive SAS RAID0 sub-system, possibly damaging the unit in the process.. More on that in a minute.

    When the SeriTek card failed, it failed suddenly and it simply went off the radar screen- disappearing from System Profiler and of course the drives would not mount. I got it back once, then it disappeared again- this time for good. I tried flushing the boot cache, re-installing the drivers, and eventually re-installing a complete system from scratch on zero-ed drives (which I did because of the problems with the ATTO card and SAS drives at the same time).



    Finally, I replaced the SeriTek card with a Newer Technology MAXPower eSATA 6G pci card, which has been working flawlessly and is apparently bootable- unlike the SeriTek card. Plus it was $39 instead of the $110 FirmTek wanted for an out-of-warranty replacement..

    I have the card connected to my backup sub-system which is a Burly 2-drive hotswap unit. In addition to replacing the pci card, I managed to finally fix a long-standing fan noise problem in the Burly (the large rear fan). Initially I used silicone spray, but that seemed to work only for a week or so, so I finally used 3-In-1 oil, and so far so good... Here's to small victories



    When the SeriTek card failed, it caused major malfunctioning of the other pci card installed- an ATTO ExpressSAS H680 8-port external SAS adapter, which is connected to a 4-drive hotswap Burly in which I have 4 x 15K.7 Seagate Cheetahs. I use this as my RAID0 Photoshop scratch disk..

    When I finally removed the failed SeriTek card from the machine, the ATTO card and attached Burly and hard drives appeared to return to normal functioning... sort of.

    I started getting write errors on one of the Cheetahs. These were reported by SoftRAID which I use as driver software for the SAS drives. SoftRAID tells you which particular drives are causing errors.

    At first I thought it was such and such particular drive, and that it had been hosed when the SeriTek card went down and took the ATTO card/drives with it. I RMA'd the "defective" drive still under warranty back to Seagate and received a replacement.. Popped it back into the unit and all was well for a little bit.. then write errors again- WTF! And on the replacement drive no less..



    Then a little birdie whispered in my ear. I pulled that drive from the #2 bay and put in a drive from a different bay which had been functioning normally. Well now that drive was getting write errors. In fact, any drive I put in the #2 bay gets write errors. In the 3 other bays, they all function normally..

    While I research this, I'm using the SAS Burly with 3 drives just fine- with nothing in the #2 bay. At this point, I am thinking the #2 bay backplane/cables is the culprit, but I suppose it could also be the cable connector... Not sure how to zero in on that just yet. Maybe, it'll be easier to send entire unit back for replace?

  3. #43
    Join Date
    Aug 2001
    Location
    Grangeville, ID USA
    Posts
    9,123

    Default

    Since that is a JBOD RAID you can go inside and swap the data cable on the back of bay 2 with bay 1. See if the problem stays with the bay or moves to the top bay with the cabling.

    I like the Newer card. We're going to carry them. We also now have a SATAIII port multiplier board to go with it, so we can get over 400 MB/sec on a single port multiplier eSATA cable.

    Rick
    molṑn labe'
    "I am a mortal enemy to arbitrary government and unlimited power. I am naturally very jealous for the rights and liberties of my country, and the least encroachment of those invaluable privileges is apt to make my blood boil."
--Ben Franklin

  4. #44
    Join Date
    Aug 2002
    Posts
    1,010

    Default

    Thank you Rick! Excellent idea- I'll try the cable swap...

  5. #45
    Join Date
    Aug 2002
    Posts
    1,010

    Default

    Did some more troubleshooting on this issue..

    I <was> pretty much settling on this being a hardware failure with the external 4-drive SAS unit..

    To make my testing a little more manageable, I removed the 3 drives formatted with SoftRAID as a RAID0 in bays 1, 3, and 4- which I thought were all performing just fine.. I previously tested this RAID with some heavy Photoshop work and no problems..

    Then I used SoftRAID to format the 4th drive as a standard drive and put it in drive bay #2. I didn't get any read/write errors testing it as Photoshop's scratch disk. I tried swapping the data cables to that drive in the #2 bay, with no difference...

    Now I'm confused because drive bay #2 seemed to be the culprit in my earlier testing..
    I moved this same drive to drive bay #1 to test, and I also decided to add a file copy routine to my testing...
    As soon as I dragged a large image file from my Desktop to this drive, the machine KP'd..

    I'm like "huh?!" wtf?

    I then tried this with this same drive in different slots, and same results: KP every time..

    Then I tried this with the SoftRAID-formatted 3-drive stripe: no problems..

    Then I broke the stripe into 3 individual volumes/derives formatted with SoftRAID: machine KP'ed every time, regardless of which drive was destination for file copy.. hmmm.

    Now I'm thinking it's not the enclosure, not the drives- it's the SoftRAID driver.

    I converted 3 of the drives to Apple DU format in SoftRAID, then did same to them in Disk Utility. Tried the same file copy test on each. No problems.
    I then made a RAID0 stripe from those 3 drives in Disk Utility and tested: no problems..

    Meanwhile, this fourth hard drive is still formatted with SoftRAID and when I attempt to convert it to Apple DU, machine immediately KP's. So I may have to send this drive back to Seagate for a replacement..

  6. #46
    Join Date
    Aug 2001
    Location
    Grangeville, ID USA
    Posts
    9,123

    Default

    Got an older OS version to test with? Hard to work on a drive that crashes your system. Bet the drive is just florfed, corrupted and not mechanical failure.

    R
    molṑn labe'
    "I am a mortal enemy to arbitrary government and unlimited power. I am naturally very jealous for the rights and liberties of my country, and the least encroachment of those invaluable privileges is apt to make my blood boil."
--Ben Franklin

  7. #47
    Join Date
    Aug 2002
    Posts
    1,010

    Default

    Yes, I agree- probably not mechanical failure..

    I've sent it back RMA to Seagate since I don't have any means to attempt a re-format with a Windows machine..

    It's situations like this that make me consider installing Windows so I could install Seagate's SeaTools and thereby probably be able to recover the drive- then bring it back to the Mac OS side..

    On the plus side- I am glad there are no apparent issues with the ATTO card and the SAS cables and enclosure..

    I think this was the second of the 2 drives that were hosed when the fried SATA card took down the SAS sub-system..

  8. #48
    Join Date
    Aug 2002
    Posts
    1,010

    Default

    My SAS RAID is back up and finally running normally..

    I finally had to replace the second "iffy" Cheetah drive, even tho it <seemed> to be mostly ok... but I was getting kernel panics when trying to use SoftRAID, so I knew something was not quite right..

    What did I learn from this?

    1. Not immediately removing the malfunctioning/failed FirmTek SATA card and trying to run the SAS RAID while troubleshooting was <NOT> a good thing to do
    Continuing to use the RAID sub-system with that failed card in place likely contributed to the malfunctioning of the PCI bus and the overall OS- which in turn set the conditions where 2 of the 4 SAS drives got whacked in a catastrophic fashion.

    2. If I had Windows installed and Seagate SeaTools drive formatting software, I probably could have recovered the two florfed Cheetahs.. As it were, I had to RMA to Seagate.

    3. SoftRAID was very quick and accurate to identify the failed Cheetahs even tho I was still able to partially recover them from time to time to the Desktop.

    So all in all, a challenging tech problem and excellent learning experience. And not too expensive!
    I'm glad I took my time getting it worked out- and thankful for the luxury that I could still run the RAID with 2 drives or not use it at all and instead rely on my fast Raptors for Photoshop work..

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •