Faulty disk? CS407e beeping and losing sda1

[jimmac]

I'm a fresh owner of a CS407e. I'm using a three Seagate ST31000333AS disk RAID-5 setup. I chose to check for bad blocks when cretaing the volume.

To have my media served to my PS3 I updated the stock firmware to DSM 2.0-0728. At about 4am I was woken up by a beeping sound and decided to turn it off not to wake my daughter. The power LED started flashing, but the beeping stopped so I thought it's flushing cache and shutting down.

This morning I found the machine inaccessible apart from responding to pings. After having to power it down, it came back with the first drive claiming being defective. Since it showed up as OK and none of the SMART info showed any sign of an issue, I started repairing the volume, which is currently in progress. Since the webUI logs don't seem to be very verbose, I looked at /var/log/messages to find the following ATA errors (apart from a flood of synoindexd errors indexing my photos). Sadly I wasn't fully awaken when the problem occurred so I'm not sure if 4AM is the time it started beeping, but could very well be.

Code: Select all

Jan  8 04:03:54 scemd: modules/raid_sys_volume_check.c:325 Md0 degraded [2/4]
Jan  8 04:13:55 kernel: ata_qc_timeout^I(3350): cannot delete scsicmd timer
Jan  8 04:13:55 kernel: ata1: command 0x25 timeout, stat 0xd0 host_stat 0x1
Jan  8 04:13:55 kernel: ata1: translated ATA stat/err 0xd0/00 to SCSI SK/ASC/ASCQ 0xb/47/00
Jan  8 04:13:55 kernel: ata1: status=0xd0 { Busy }
Jan  8 04:13:55 kernel: end_request: I/O error, dev sda, sector 5062966
Jan  8 04:13:55 kernel: raid1: sda2: rescheduling sector 82816
Jan  8 04:13:55 kernel: ATA: abnormal status 0xD0 on port 0xC5000E87
Jan  8 04:13:55 kernel: ATA: abnormal status 0xD0 on port 0xC5000E87
Jan  8 04:13:56 kernel: ATA: abnormal status 0xD0 on port 0xC5000E87


Is there a way to tell if this is a HW issue or a kernel issue for that particular firmware release?

[jimmac]

The symptoms in thread http://www.synology.com/enu/forum/viewtopic.php?f=7&t=10599 seems to match mine, will try replacing the cable.

[Franklin]

Greetings jimmac

The ST31000333AS or the ST31000340AS is not compatible with x07 models, so you may be experiencing a disk compatibility issue. Please consider using the WD10EACS for your system.

However, if you must use the ST31000333AS drive...your error is similar to an error experienced by the ST31500341AS, and it was only resolved by updating the firmware of the HDD by Seagate.

Another thought, please also make sure you are using firmware 803 with your system.

Hope this helps

[jimmac]

Thanks for the info.

I haven't checked the disk firmware yet, but after replacing the long sata cable for disk1 I no longer get any problems. Granted a week isn't a long time to be relevant. I'll keep people posted if anything happens. For now I'm blaming the cable

[jimmac]

It wasn't the cable :/ Just had another failure of the first disk.

Code: Select all

Jan 17 12:46:12 kernel: ata_qc_timeout^I(3350): cannot delete scsicmd timer
Jan 17 12:46:12 kernel: ata1: command 0x35 timeout, stat 0xd0 host_stat 0x1
Jan 17 12:46:12 kernel: ata1: translated ATA stat/err 0xd0/00 to SCSI SK/ASC/ASC
Q 0xb/47/00
Jan 17 12:46:12 kernel: ata1: status=0xd0 { Busy }
Jan 17 12:46:12 kernel: end_request: I/O error, dev sda, sector 4979903
Jan 17 12:46:12 kernel: md: super_written gets error=-5, uptodate=0, dev=sda1
Jan 17 12:46:12 kernel: raid1: Disk failure on sda1, disabling device.
Jan 17 12:46:12 kernel: ^IOperation continuing on 2 devices
Jan 17 12:46:12 kernel: ATA: abnormal status 0xD0 on port 0xC5000E87
Jan 17 12:46:12 kernel: ATA: abnormal status 0xD0 on port 0xC5000E87
Jan 17 12:46:12 kernel: ATA: abnormal status 0xD0 on port 0xC5000E87



Franklin, when you mentioned 803 firmware I didn't realize you mean the actual cubestation firmware still in beta. Any estimates when it's out of beta? Don't want to end up regretting the upgrade if there's no downgrade path. Especially since I'm not certain it's not actually the disk being wonky as it's not officially supported.

[Franklin]

@jimmac

You may elect to use firmware 731 which was just released today, which does have enhancements to HDD error handling, this firmware may help you, and its a current production firmware.

Also, please refer to the Seagate firmware notice - that may also help you as well.

Please note that the Seagate .11 compatibility issue still exists for all x07 models - not matter which firmware is used on either the Synology system or the Seagate HDD - it is a hardware incompatibility.

Hope this helps

[jimmac]

Thanks for the heads up, Franklin. I wish I had waited for the WD drives to become available and not order an unsupported Seagate drive . So much for going with the more expensive solution. Hoping for the best.

[jimmac]

After downgrading to a two disk RAID1 and disabling the hybernation, the system seems stable.

[Rude]

The Seagates drives should be fine after upgrading them with the proper firmware. I'm running 3 Seagate .11 (1.5TB) drives with the proper firmware without problems. I did have a problem with my CS407e freezing but that doesn't seem to be HDD related (more download manager related) If it did I would have lost the volume but everything is still running fine.

I'm retracting the above post. The new firmware is NO guarantee that the drives will run stable in a RAID5 setup! There seems to be something else going on with some of the 7.11 series.