March 5, 201313 yr So, my friend has been having issues with his unraid AGAIN... he shut it down for about a month while he was doing basement renovations, and when he turned it on last week, it ran for about an hour, and then took out his network (his words) He brought it over, and I took a look at it. the first 2 times i powered it on, it didn't really do anything... then i got it to post, and boot... wasn't reachable by the network, and the keyboard was working, but when it finished booting, it didn't. hard reset. got access to the server via web, and telnet... and keyboard was working again. of the 5 data, 1 parity and 1 cache drives, 2 data drives were not shown... upon rebooting a few more times... i could get it to where 1 drive was missing... but never had all drives shown. When I did `fdisk -l | grep "/dev"` (so many lines shown without the grep), it shows that all the drives are there. Any ideas what I should be looking for/at? The server is running 5.0 RC8, and it was running fine? At first I thought maybe power supply was flaky, but since fdisk is showing that everything is there? I am lost... please help!
March 5, 201313 yr You mentioned that the machine was shut down while renovation was going on which suggests that it could have been moved around quite a bit so that the internals got disturbed. I would check that any daughter boards are properly seated and then check that all the cabling appears to be securely plugged in. Also check that there is not a lot of dust inside the system (in case it was not protected from this during renovation). The fact that you were having trouble even getting the system to boot is a bit suspicious.
March 5, 201313 yr Author I have done that, i have reseated the raid card, triple checked all the sata and power connectors... and there is no dust inside the case, and i know 100% for a fact, it was not jostled around, or moved around (was put into another room out of harms way on purpose)... he is pretty anal about his computer stuff. are you thinking maybe power issues as well? (ie the power supply)
March 5, 201313 yr Not sure. Simply guessing as the system sounds unhappy at a level beyond the software level. The syslog might give an indication of what is happening under the covers. It will at least show what the Linux level thinks of the drives.
March 5, 201313 yr Author I'll turn it back on tonight, and post the results. meant to grab it last night, but forgot
March 5, 201313 yr chkdsk the flash drive. Can I ask why you gave this advice? I find it hard to believe that a program would be corrupt, yet work for all disks except one. I'd love to expand my knowledge, this isn't me being an ass.
March 5, 201313 yr chkdsk the flash drive. Can I ask why you gave this advice? I find it hard to believe that a program would be corrupt, yet work for all disks except one. I'd love to expand my knowledge, this isn't me being an ass. he mentioned drives would sometimes appear, sometimes not appear, etc during boot. my experience in the forums and with unraid is that a error on the flash drive (usb stick) can cause really funky things to happen. its an easy check to perform.
March 6, 201313 yr Author Ok, checkdsk didn't find anything... I had actually had all drives shown on my second last boot... then when i started the array (said it wasn't started due to improper shutdown) it started...a nd then hung... so, reboot... 1 drive not started... I copied the dmesg and the syslog... here is the syslog: http://pastebin.com/JP9tMvtR looks like I have 1, possibly 2 drives that are failing/failed, which would explain why sometimes i have all, or missing up to 2 drives when i boot up... [some of] the errors are: Mar 5 18:50:08 Tower kernel: sd 1:0:3:0: [sde] Unhandled error code Mar 5 18:50:08 Tower kernel: sd 1:0:3:0: [sde] Result: hostbyte=0x01 driverbyte=0x00 Mar 5 18:50:08 Tower kernel: sd 1:0:3:0: [sde] CDB: cdb[0]=0x28: 28 00 00 00 10 3f 00 00 08 00 Mar 5 18:50:08 Tower kernel: end_request: I/O error, dev sde, sector 4159 Mar 5 18:50:08 Tower kernel: Buffer I/O error on device sde1, logical block 4096 Mar 5 18:50:08 Tower kernel: Buffer I/O error on device sde1, logical block 4097 Mar 5 18:50:08 Tower kernel: Buffer I/O error on device sde1, logical block 4098 Mar 5 18:50:08 Tower kernel: Buffer I/O error on device sde1, logical block 4099 Mar 5 18:50:08 Tower kernel: Buffer I/O error on device sde1, logical block 4100 Mar 5 18:50:08 Tower kernel: Buffer I/O error on device sde1, logical block 4101 Mar 5 18:50:08 Tower kernel: Buffer I/O error on device sde1, logical block 4102 Mar 5 18:50:08 Tower kernel: Buffer I/O error on device sde1, logical block 4103 Mar 5 18:50:08 Tower kernel: sd 1:0:3:0: [sde] Synchronizing SCSI cache Mar 5 18:50:08 Tower kernel: sd 1:0:3:0: [sde] Result: hostbyte=0x01 driverbyte=0x00 Am I right to think that drives are failed/failing? and I should prolly pull them out and recover (at least 1 of) them manually?
Archived
This topic is now archived and is closed to further replies.