In an effort to reduce heat and power, and increase capacity, I'm gradually replacing my Seagate and Hitachi 750 GB dries with WD 1 TB Green Power drives. I did such a replacement last night. Rebooting my server, and I got error after error from the console. Finally after 5 minutes I got a login prompt. Trying to connect via web browser and there was no sign of the server.

ifconfig showed that it wasn't connected either.

The new drive was connected to my Promise TX4 controller. On startup, the controller recognised three of the four disks, but the fourth (the new one) had a non standard  hard drive entry instead of WDxxxxxxx, it was P02  xx or something similar. After trying to reboot multiple times via console, I did manage to finally connect to the server via browser. None of the disks were found and the array didn't come online of course.

panicking somewhat, I shutdown, disconnected the new drive and reconnected my Hitachi, without installing it into a drive cage. This time the server started and the array was up in no time. Suspecting the new 1 TB disk, then, I replaced the Hitachi with yet another 1 TB WD Green Power disk that I had recently bought. This time all was well again, and unRAID correctly recognised that I had replaced a disk. The data was rebuilt overnight, and was 20 minutes from completion when I left for work this morning.

 

So, assuming that I have a bad disk, is the behaviour of unRAID normal- taking forever to start up, not recognising any of the disks once started, and only making the server connectible once in about ten attempts? Any insights will be appreciated.

So, assuming that I have a bad disk, is the behaviour of unRAID normal- taking forever to start up, not recognising any of the disks once started, and only making the server connectible once in about ten attempts? Any insights will be appreciated.

There is no way to answer your question.  It is like asking how long it would take to unlock your car using your keychain remote if an intermittant short were to occur in its wiring.  You might need to try a few times.  Each error (and there were probably many thousands of errors) caused the OS to try to handle it, and to log it.  If it runs out  of log space, and it kills processes to free RAM, you do not get to log in.  It is a race to use the available resources.

 

There is no "normal" behavior when any PC is "not-normal."  It might be completely dead, or ignore the new hardware, or limp along with errors every time it attempts to access the new disk, or if that disk is interfering with other hardware, you might get errors when it attempts to access anything.  You are lucky you knew the hardware you had just installed was the most likely suspect.

 

The clues as to what was happening were on the system console, and possibly in the system log.  That is the normal behavior.    If the bad drive was causing thousands of errors per second, then it might take a while to get you logged in.

 

Glad it is working with the replacement disk.

 

Next time, it would help to know more about the errors you were seeing on the console.  (take a picture with a camera if there is no other way to get a copy)

 

Joe L.

Thanks for the response, Joe. I figured I'd ask the vague question in case it was something that has been seen with frequency, but wasn't pulled up by a search.

All I can say is that I'm glad all is okay right now, and that it's a pain to try and replace the disk RMA and all that. I am going to try my USB to SATA adapter on the disk tonight to see if it'll work okay in Windows.

If it doesn't work there, that's the end of the matter and I don't need to take console shots or anything in unRAID; it goes back to WD. My server works fine with an identical model disk so that's it. If it does work in Windows, on the other hand, I may well take a look at syslogs and take console screenshots in an effort to be more helpful when asking for help.

Cheers.