August 19, 201015 yr It seems that Seagate is limiting firmware CC35 to only a sub-selection of specific serial number drives [ http://forums.seagate.com/t5/Barracuda-XT-Barracuda-and/Idle-clicking-of-Seagate-Barracuda-LP-HDs-and-DOA-issues/m-p/51064#M19467 ] You can force an upgrade: http://niallbest.com/seagate-2tb-st32000542as-cc35-firmware-upgrade/ http://stx.lithium.com/t5/Barracuda-XT-Barracuda-and/Idle-clicking-of-Seagate-Barracuda-LP-HDs-and-DOA-issues/m-p/55298#M20418 This thread finally got me motivated tonight. I updated all 12 of the drives in my server. Thankfully they all went without a hitch, parity check running right now. Serial numbers ranged from 5X-9X and firmware numbers from CC32-CC34. I was only able to use the onboard intel sata chipset (it was in AHCI mode). The onboard JMicro ports and the adaptec 1430 did not work.
August 19, 201015 yr Thanks for posting it. I found that thread and post when I started researching but hadn't gotten around to posting it here. I got a bit distracted with running some preliminary tests. I'm in process of doing a preclear now. I was curious to see how various regions of the drive performed speed wise and how it compares to my current 1TB WD FALS cache/boot/root/swap drive. I partitioned it roughly in half then setup another partition for Linux boot/root, and one for swap. Then I ran a simple diskspeed script to test speed. For grins, I even tested some ancient 10K RPM WD Raptor 74GB SATA drives. It's not exactly apples-to-apples, as the 1TB is 37% and 10% full on the equiv ReiserFS and Ext3 portions, while the 2TB Seagate is empty and the CPUs are different, but system memory is identical. Seagate 2TB SATA 5900 RPM ST32000542AS sdb1 Primary Linux ReiserFS 980741.27 sdb2 Primary Linux ReiserFS 980741.27 sdb3 Primary Linux ext3 30721.43 sdb4 Primary Linux swap 8192.38 /mnt/tmp/sdb1# ~/diskspeed 4096000000 bytes (4.1 GB) copied, 40.8918 s, 100 MB/s 4096000000 bytes (4.1 GB) copied, 42.6702 s, 96.0 MB/s 4096000000 bytes (4.1 GB) copied, 34.6182 s, 118 MB/s /mnt/tmp/sdb2# ~/diskspeed 4096000000 bytes (4.1 GB) copied, 49.2317 s, 83.2 MB/s 4096000000 bytes (4.1 GB) copied, 47.7500 s, 85.8 MB/s /mnt/tmp/sdb3# ~/diskspeed 4096000000 bytes (4.1 GB) copied, 73.6467 s, 55.6 MB/s 4096000000 bytes (4.1 GB) copied, 72.5483 s, 56.5 MB/s WD 1TB 7200 RPM WD1001FALS-00J7B0 sda1 Primary Linux ReiserFS 846603.40 sda2 Primary NTFS 40961.58 sda3 Primary Linux swap 10240.48 sda4 Primary Linux ext3 102396.52 /mnt/tmp/sda1# ~/diskspeed 4096000000 bytes (4.1 GB) copied, 49.6407 s, 82.5 MB/s 4096000000 bytes (4.1 GB) copied, 49.5994 s, 82.6 MB/s /mnt/tmp/sda4# ~/diskspeed 4096000000 bytes (4.1 GB) copied, 79.4802 s, 51.5 MB/s 4096000000 bytes (4.1 GB) copied, 77.8310 s, 52.6 MB/s WD 74GB SATA 10K RPM WD740GD-00FLA1 sdb1 Primary Linux ReiserFS 35434.51 sdb2 Primary Linux ext3 30721.43 sdb3 Primary Linux swap 8192.38 /dev/sdb1# ~/diskspeed 4096000000 bytes (4.1 GB) copied, 51.9476 s, 78.8 MB/s 4096000000 bytes (4.1 GB) copied, 52.1988 s, 78.5 MB/s /dev/sdb2# ~/diskspeed 4096000000 bytes (4.1 GB) copied, 54.9703 s, 74.5 MB/s 4096000000 bytes (4.1 GB) copied, 55.1367 s, 74.3 MB/s
August 19, 201015 yr It's not exactly apples-to-apples, as the 1TB is 37% and 10% full on the equiv ReiserFS and Ext3 portions, while the 2TB Seagate is empty and the CPUs are different, but system memory is identical. Comparing reiserfs to ext3 skews things a bit. I don't think reiserfs has inode tables the same way ext3 does.
August 19, 201015 yr I had the same issue with the boot disk, so I just kept the firmware and precleared it. No issues at all. It precleared on par with my other 2TB drives, maybe just a little faster. Seems fine to me. Alright, so I burned the boot image (CD ISO file), hooked up the optical drive, boot from newly burned media into their tool and ran into a few issues. The first issue is their scan would only detect the first drive which was a WD and would not scan further. I shutdown poweroff, disconnect all but the Seagate drive and reboot. With only the Seagate connected, it would not detect any drives at all. I've seen this behavior before, so I had an idea of what to try. I reboot into bios, flip the SATA mode from AHCI to EIDE. I try the scan, it finds the drive, so I tell it to download firmware. It takes a few seconds then kicks back "Model Matched. No Firmware Match! Will Not Download Firmware! -- Model matched but firmware version is not compatible." Now this is getting really annoying. It seems that Seagate is limiting firmware CC35 to only a sub-selection of specific serial number drives [ http://forums.seagate.com/t5/Barracuda-XT-Barracuda-and/Idle-clicking-of-Seagate-Barracuda-LP-HDs-and-DOA-issues/m-p/51064#M19467 ]
August 19, 201015 yr Comparing reiserfs to ext3 skews things a bit. I don't think reiserfs has inode tables the same way ext3 does. Yeah, it does change the numbers a little bit. For whatever reason, I've always used EXT3 as the /boot filesystem, so I was curious what the drive speeds were in similar situation. That made it more of a Disk to Disk with a hint of beginning and end of the drive comparisons. So far I'm liking this 5900 RPM 2TB Seagate drive. It finished the preclear single pass process without issue and seems to be on par or better than my older 7200 RPM 1TB WD drive.
August 20, 201015 yr Just an FYI, my sale drive came in today. I had to force an update, unlike the others that auto-updated.
August 20, 201015 yr I had the same issue with the boot disk, so I just kept the firmware and precleared it. No issues at all. It precleared on par with my other 2TB drives, maybe just a little faster. Seems fine to me. Run a SMART report and take a look at the Seek_Error_Rate and/or Raw_Read_Error_Rate...
August 20, 201015 yr Well I'll be damned.. You're right Neil. I guess I'll be taking the tower down this weekend and forcing the flash. Thanks for the info. I had the same issue with the boot disk, so I just kept the firmware and precleared it. No issues at all. It precleared on par with my other 2TB drives, maybe just a little faster. Seems fine to me. Run a SMART report and take a look at the Seek_Error_Rate and/or Raw_Read_Error_Rate...
August 20, 201015 yr Author I had the same issue with the boot disk, so I just kept the firmware and precleared it. No issues at all. It precleared on par with my other 2TB drives, maybe just a little faster. Seems fine to me. Run a SMART report and take a look at the Seek_Error_Rate and/or Raw_Read_Error_Rate... Did the same thing as dogmanky. Currently 3% into Post-Read step. So I took a look at the smart_start report and yes, tons for Seek_Error_Rate and Raw_Read_Error_Rate Guess I need to force the flash also. Question, would the preclear still be good/valid after forcing the flash? or would I need to do it again?
August 20, 201015 yr The firmware update is non-destructive for the data stored on the drive. Also, when looking at the smart report for those attributes, you need to focus on the Current Value, Worst Value, and Threshold Value. I do not believe the raw value (last column) produces any meaningful information for us.
August 20, 201015 yr The firmware update is non-destructive for the data stored on the drive. that's awesome news! I'm a few percent away from wrapping up my preclear. I'm glad I don't need to wait for another 24hrs to run that again I guess it's time to create the firmware boot disc...
August 20, 201015 yr Seagates are known for crazy Raw_Read_Error_Rate and Seek_Error_Rate counts. My oldest has a seek error rate of... 38715245524.
August 20, 201015 yr Author Agreed. Great news. Still playing around with setup/configuration/running cables so no rush, but glad to know I don't need to rerun the preclear. Also good to know to not pay attention to the raw numbers.
August 20, 201015 yr Also good to know to not pay attention to the raw numbers. Do pay attention to these two raw numbers, though: 197 Current_Pending_Sector 198 Offline_Uncorrectable
August 20, 201015 yr I just upgraded mine (forced) to CC35. I think the main reason for the firmware release is to stop it from clicking and other bad things -- I haven't experienced any problems with my drive, but it seems prudent to upgrade, given that Seagate have said it's OK to force it. I can confirm it's non-destructive. I backed up beforehand, but the data is still intact. I've kept a record of the smartctl report and will keep an eye on the drive to see if the Raw_Read_Error_Rate and Seek_Error_Rate numbers change.
August 21, 201015 yr I just forced mine to CC35. Not sure what the hell all those switches in the cmd line did but it was quick and painless. Powered back up and all is well.
August 21, 201015 yr Raw_Read_Error_Rate and Seek_Error_Rate are sector counts and seek counts on Seagate drives, not error counts. For an explantion see the following posts. http://groups.google.com/group/microsoft.public.windowsxp.help_and_support/msg/71e9ab48939aab6a http://groups.google.com/group/microsoft.public.windowsxp.help_and_support/msg/597c2ecc5ff526d9
August 21, 201015 yr Raw_Read_Error_Rate and Seek_Error_Rate are sector counts and seek counts on Seagate drives, not error counts. That's misleading. Thanks, Seagate! Another post I read made it easier for me to follow: "Did you notice that your normalised value (68) isn’t even close to the threshold (30)? I think your disk was fine. The man page says if normalised drops below threshold you’re in trouble."
August 22, 201015 yr Can I run this update with the drive in a usb dock on my main computer or is it best to install it in a sata port to run the update?
August 22, 201015 yr Can I run this update with the drive in a usb dock on my main computer or is it best to install it in a sata port to run the update? I tried running the "Windows" version with the drives in an external docking station via a PCMIA SATA Controller and a USB docking station. It was a no go. The "Windows" version is only a window apps to display the warnings and disclosures, then it reboots the computer into what looks like a DOS environment to actually perform the update. Hence it never load the necessary drivers to access external devices. I even tried connecting the HD docking station to a SYBA Silicon 3124 4 port SATA Controller (2 eSata + 2 internal). The "Windows" flasher identified all of the Seagates as non-Seagate drives. I will try again with the bootable CD.
August 22, 201015 yr Seagate recomends running the sata ports are in IDE mode, sometimes called legacy mode or compatibility mode. It also says make sure no other HDDs are attached. Both of which recomendations I would follow.
August 27, 201015 yr I forced it following the instructions here. I unhooked my other Seagate drive but left my OCZ SSD plugged in - the firmware updater ignored it. I did have to set all the drives to IDE mode as the guide says. Well, that was definitely a pain, plus a waste of a CD. Sure hope it was worth it. I didn't see any error messages, but I'm still curious - how can I check that the firmware update actually worked?
August 27, 201015 yr hdparm -i /dev/sdX Model=ST32000542AS, FwRev=CC35, SerialNo=6XW1PDHJ Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
August 27, 201015 yr Crap, mine is still on CC34. SpeedFan on Windows confirms this. SpeedFan also reports some SMART errors: Attribute Current Raw Overall Reported Uncorrectable Errors 99 1 Watch Warning: Reported Uncorrectable Errors is below the average limits (100-100). Hardware ECC Recovered 29 30442515 Watch Warning: Hardware ECC Recovered is below the average limits (34-100). Weaksauce. I'm regretting buying this drive now. Edit: I tried forcing it again, but this time making sure it was the only drive connected (no OCZ SSD this time). Same result. I seem to be stuck on CC34. SpeedFan also doesn't recognize the drive after this second attempt, but unRAID still does. Oh well, guess I'm giving up for now. I'm adding this drive as my parity drive. If it gives me any trouble I guess I'll just return it (as opposed to RMAing it).
Archived
This topic is now archived and is closed to further replies.