I received a message that my node was offline. The service would not start because the disk is missing. when I reboot the pc disk management says I need to initialize the disk (which I did not do). Does this mean the disk died? The node is only 3 months old with just 352GB used space. The other node on this PC is working fine. There is a copy of the missing disk node data from early September. If I repoint the node to that data will it get DQed?
Your suggestion as to how to proceed is appreciated.
What kind of disk is this? I have seen similar issues with iSCSI targets, in which case chkdsk was able to make data visible again.
Otherwise you can try power-cycling everything. It’s very unlikely that the disk died in a way that ti forgot everything but otherwise pretends to be OK. It likely some issue in-between.
Yes.
1 Like
It is an old WD Caviar Blue 2TB CMR disk. I can not run chkdsk because it has no drive letter. Not sure what you mean by power-cycling everything. I already tried a reboot of the pc. That is when it said to initialize the disk.
Turn power OFF, wait 5 min, and then turn power on. Reboot does not remove power from the drive. It may fix some issues if e.g. driver controller hung.
If it still does not show any data you may want to try some data recovery apps, that may be able to recover partition table. NTFS is quite hard to murder, it keeps two copies of MFT.
2 Likes
Thanks you. That worked. Disk is there, service is running and dashboard is working.
3 Likes
You can also unplug the system from the wall and press the power button a few times. It will drain the capacitors quicker.
But I learned something new today too. I didn’t knew the power cycling can make such difference like reviving a drive.
4 Likes
I had this issue when I used a USB drive. I plugged it to a different port and it came back, the node was starting without issue. After some time I had to power cycle the PC as I run out of ports. 
My best guessing was that the system was not able to handle all the writes and run out of buffer. As I moved that node from external drive to internal, it did not happen again.
1 Like
And this is for a reason
Sometimes I even agree with
It may be useful. Have had this experience many times when I was working as a System Administrator on demand.