May 8, 201115 yr Bottom Line Up Front - I am trying to add my old parity drive in as a new data drive and after I tell it to format the drive, it returns and still shows it is unformatted. I've attached the syslog. I've tried 3 times in 2 different slots in the IcyDock on different SATA ports. Background - my parity drive was occasionally disappearing (would show up blue) and I'd have to rebuild. Replaced the cable for that drive and it seemed to clear it up. I was also running out of space so I picked up a new Hitachi 3TB drive (previous parity was Seagate 1.5TB drive), used the forum instructions and added it as a 2.2 TB parity. More Info - I wanted to add the old 1.5TB parity drive as a new data drive. I ran it through the preclear script without the extra read at the end. It seemed to complete ok, although at the end it said there were differences in the SMART statistics, specifically high-fly writes. I ran an extended SMART test on it and it passed, so I stuck the drive in, added it to the array, it showed up unformatted and I told it to format. After a while it came back and showed that it was still unformatted. The log showed similar to this: May 8 20:25:03 WatchTower logger: mount: wrong fs type, bad option, bad superblock on /dev/md5, May 8 20:25:03 WatchTower logger: missing codepage or helper program, or other error May 8 20:25:03 WatchTower logger: In some cases useful info is found in syslog - try May 8 20:25:03 WatchTower logger: dmesg | tail or so May 8 20:25:03 WatchTower logger: May 8 20:25:03 WatchTower emhttp: _shcmd: shcmd (126): exit status: 32 May 8 20:25:03 WatchTower emhttp: disk5 mount error: 32 May 8 20:25:03 WatchTower emhttp: shcmd (127): rmdir /mnt/disk5 I thought maybe since I took a shortcut on the preclear it had messed up something, so I started from scratch with a full preclear. Everything above then repeated including new high fly writes and it still would not format. Thinking it might be a hardware problem, I put it in a different slot in the IcyDock and hooked it up to the SATA port that my cache drive has been running fine on. I started it up without running preclear on it this time, so when I added it to the array it started by clearing it. That seemed to complete fine, it showed unformatted so I told it to format, and after a while it came back as unformatted again. I've attached the log. I'm at a loss at what to do now. Is this a problem with the hard drive or what? syslog-2011-05-08.txt
May 9, 201115 yr Author I ran a reiserfsck, got the following: reiserfs_open: the reiserfs superblock cannot be found on /dev/sdh1. Failed to open the filesystem. If the partition table has not been changed, and the partition is valid and it really contains a reiserfs partition, then the superblock is corrupted and you need to run this utility with --rebuild-sb. I tried to format the drive myself from the command line (mkreiserfs -q /dev/md5) to see what it would say and it seemed to complete without any errors, but reiserfsck gives the same error.
May 9, 201115 yr Author I could still use some help with this, I just ran memtest for 8 hours with no errors. Not sure what else could be the problem. I'm going to run another extended SMART test but the last one passed.
May 10, 201115 yr I only skimmed your post, but I had a similar problem. I rebooted and I could then successfully format the drive.
May 10, 201115 yr Author I have rebooted several times and tried the drive in different slots in my IcyDock. This morning I pulled the drive and stuck it in an eSata dock on my Win7 desktop and formatted it NTFS. Then I put it back into my unRaid and told it to format it again. It did seem to format this time with the following errors in the log which I am guessing are not very good. May 10 13:44:17 WatchTower emhttp: shcmd (59): mkreiserfs -q /dev/md5 2>&1 | logger May 10 13:44:17 WatchTower logger: mkreiserfs 3.6.21 (2009 www.namesys.com) May 10 13:44:17 WatchTower logger: May 10 13:44:17 WatchTower kernel: attempt to access beyond end of device May 10 13:44:17 WatchTower kernel: sdh1: rw=0, want=2930276992, limit=2930272256 May 10 13:44:17 WatchTower kernel: md: disk5 read error May 10 13:44:17 WatchTower kernel: handle_stripe read error: 2930276984/2, count: 1 May 10 13:44:17 WatchTower kernel: attempt to access beyond end of device May 10 13:44:17 WatchTower kernel: sdh1: rw=1, want=2930276992, limit=2930272256 May 10 13:44:17 WatchTower kernel: md: disk5 write error May 10 13:44:17 WatchTower kernel: handle_stripe write error: 2930276984/2, count: 1 May 10 13:44:17 WatchTower kernel: md: recovery thread woken up ... May 10 13:44:17 WatchTower kernel: md: recovery thread has nothing to resync May 10 13:44:21 WatchTower exiting on signal 15 May 10 13:44:22 WatchTower syslogd 1.4.1: restart. May 10 13:44:22 WatchTower kernel: klogd 1.4.1, log source = /proc/kmsg started. May 10 13:44:22 WatchTower unmenu-status: Starting unmenu web-server May 10 13:44:23 WatchTower apcupsd[2248]: apcupsd exiting, signal 15 May 10 13:44:23 WatchTower apcupsd[2248]: apcupsd shutdown succeeded May 10 13:44:25 WatchTower apcupsd[2513]: apcupsd 3.14.8 (16 January 2010) slackware startup succeeded May 10 13:44:26 WatchTower apcupsd[2513]: NIS server startup succeeded May 10 13:44:31 WatchTower init: Re-reading inittab May 10 13:46:15 WatchTower emhttp: shcmd (60): mkdir /mnt/disk5 May 10 13:46:15 WatchTower emhttp: shcmd (61): set -o pipefail ; mount -t reiserfs -o noacl,nouser_xattr,noatime,nodiratime /dev/md5 /mnt/disk5 2>&1 | logger May 10 13:46:15 WatchTower kernel: REISERFS (device md5): found reiserfs format "3.6" with standard journal May 10 13:46:15 WatchTower kernel: REISERFS (device md5): using ordered data mode May 10 13:46:15 WatchTower kernel: REISERFS (device md5): journal params: device md5, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30 May 10 13:46:15 WatchTower kernel: REISERFS (device md5): checking transaction log (md5) May 10 13:46:16 WatchTower kernel: REISERFS (device md5): Using r5 hash to sort names May 10 13:46:16 WatchTower kernel: REISERFS (device md5): Created .reiserfs_priv - reserved for xattr storage. May 10 13:46:16 WatchTower emhttp: resized: /mnt/disk5 May 10 13:46:16 WatchTower emhttp: shcmd (62): chmod 700 '/mnt/disk5' May 10 13:46:17 WatchTower emhttp: shcmd (63): rm /etc/samba/smb-shares.conf >/dev/null 2>&1 May 10 13:46:17 WatchTower emhttp: shcmd (64): cp /etc/exports- /etc/exports May 10 13:46:17 WatchTower emhttp: shcmd (65): killall -HUP smbd May 10 13:46:17 WatchTower emhttp: shcmd (66): /etc/rc.d/rc.nfsd restart | logger I did run another extended SMART test last night that also passed. Anyways, now that it seems to be formatted, it is still showing red, but the drive is active and I can put files directly on it. I guess I will reinit the array and rebuild parity with it. I'm reluctant to do so if the drive is questionable though. I guess I can try filling the disk and seeing if I see any more read/write errors. Although the SMART test passed, some of the values did change, if anyone can help me decipher if I should be concerned or not I'd appreciate it: Before last extended smart: SMART Attributes Data Structure revision number: 10 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x000f 114 099 006 Pre-fail Always - 68560119 3 Spin_Up_Time 0x0003 095 090 000 Pre-fail Always - 0 4 Start_Stop_Count 0x0032 098 098 020 Old_age Always - 2097 5 Reallocated_Sector_Ct 0x0033 045 045 036 Pre-fail Always - 2285 7 Seek_Error_Rate 0x000f 076 060 030 Pre-fail Always - 44665790 9 Power_On_Hours 0x0032 078 078 000 Old_age Always - 19572 10 Spin_Retry_Count 0x0013 100 100 097 Pre-fail Always - 6 12 Power_Cycle_Count 0x0032 100 037 020 Old_age Always - 163 184 End-to-End_Error 0x0032 100 100 099 Old_age Always - 0 187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0 188 Command_Timeout 0x0032 100 097 000 Old_age Always - 14 189 High_Fly_Writes 0x003a 001 001 000 Old_age Always - 1070 190 Airflow_Temperature_Cel 0x0022 063 052 045 Old_age Always - 37 (Lifetime Min/Max 27/39) 194 Temperature_Celsius 0x0022 037 048 000 Old_age Always - 37 (0 17 0 0) 195 Hardware_ECC_Recovered 0x001a 039 022 000 Old_age Always - 68560119 197 Current_Pending_Sector 0x0012 100 100 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0010 100 100 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 14 After extended SMART: SMART Attributes Data Structure revision number: 10 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x000f 114 099 006 Pre-fail Always - 68607776 3 Spin_Up_Time 0x0003 092 090 000 Pre-fail Always - 0 4 Start_Stop_Count 0x0032 098 098 020 Old_age Always - 2100 5 Reallocated_Sector_Ct 0x0033 045 045 036 Pre-fail Always - 2285 7 Seek_Error_Rate 0x000f 076 060 030 Pre-fail Always - 44680982 9 Power_On_Hours 0x0032 078 078 000 Old_age Always - 19585 10 Spin_Retry_Count 0x0013 100 100 097 Pre-fail Always - 6 12 Power_Cycle_Count 0x0032 100 037 020 Old_age Always - 165 184 End-to-End_Error 0x0032 100 100 099 Old_age Always - 0 187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0 188 Command_Timeout 0x0032 100 097 000 Old_age Always - 14 189 High_Fly_Writes 0x003a 001 001 000 Old_age Always - 1070 190 Airflow_Temperature_Cel 0x0022 064 052 045 Old_age Always - 36 (Lifetime Min/Max 36/37) 194 Temperature_Celsius 0x0022 036 048 000 Old_age Always - 36 (0 17 0 0) 195 Hardware_ECC_Recovered 0x001a 035 022 000 Old_age Always - 68607776 197 Current_Pending_Sector 0x0012 100 100 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0010 100 100 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 14
May 10, 201115 yr Author Ok I think this drive is toast. I just did an initconfig and started rebuilding parity and the drive immediately went back to showing unformatted with this in the syslog: May 10 14:09:21 WatchTower logger: mount: wrong fs type, bad option, bad superblock on /dev/md5, May 10 14:09:21 WatchTower logger: missing codepage or helper program, or other error May 10 14:09:21 WatchTower logger: In some cases useful info is found in syslog - try May 10 14:09:21 WatchTower logger: dmesg | tail or so May 10 14:09:21 WatchTower logger: May 10 14:09:21 WatchTower emhttp: _shcmd: shcmd (125): exit status: 32 May 10 14:09:21 WatchTower emhttp: disk5 mount error: 32 May 10 14:09:21 WatchTower emhttp: shcmd (126): rmdir /mnt/disk5 I don't understand how a drive I've used this long as a parity drive suddenly flunks out when I try to switch it to a data drive, but I'll go ahead and try to RMA it.
May 10, 201115 yr You could also try running the manufacturer's test tool. SMART may not pick up certain problems.
May 10, 201115 yr Author That's what I'm doing now, Seagate requires it for an RMA. So far the test tool is not working as expected. For instance, it does not show the serial number for the Seagate drive even though it shows it for my OCZ Vertex 2 SSD. Most of the tests I am trying to run show "Test Unavailable" when I try to start them, although it did let me run a "short generic" test which passed, and I am about to run a "long generic" test. "Advanced Tests" are unavailable. I'm concerned it won't give me an RMA code even though the drive is definitely wonky since it won't let me do all the testing. Ugh, what a nightmare.
May 10, 201115 yr Seagate codes for when you don't get a code: http://sites.google.com/site/tejasharia/seagate-warranty-codes :-)
May 10, 201115 yr That "attempt to access beyond end of device" error sure looks like a HPA problem but I only see 1 HPA which appears to be for the 3T drive. Peter
May 10, 201115 yr Author I did encounter an HPA problem when I first switched to this MB (Gigabyte). I booted up one time and it told me the parity drive was not the largest drive (any more - since it had been working). I followed some instructions on the forum on how to clear the HPA and I disabled the option in the BIOS that was creating it. I've been in the BIOS while troubleshooting this issue and can confirm that HPA is still disabled. But now I'm concerned that maybe the HPA repair I did was somehow only partially successful?
May 10, 201115 yr Author I just used SeaTools to resize the drive to 32GB then back to max native size (power cycling between). From what I understand, that should clear any HPAs. I'm going to put it back in my unRaid and preclear it again and give it one more shot. My backup drive was just received by Samsung RMA so the replacement should be on its way, once I get it (assuming that one works as expected) I should be able to definitively rule out non-drive problems. Unfortunately it's coming cross country so I probably won't get it until next week.
May 11, 201115 yr Author Still can't get it to add, same symptoms, it says it is unformatted, and after the format it still shows unformatted. High_fly_writes increased again during the preclear. I'll try to RMA it with one of the error codes that Neil posted. I don't trust this drive with my data. Thanks for all the suggestions, looks like it was just a flaky drive.
Archived
This topic is now archived and is closed to further replies.