Notifications
Clear all
Topic starter
Hi, I am a newbie with SoftRaid and working with RAID systems in general, so bear with me.
Our RAID5 volume was running well until two disks failed simultaneously (same batch, of course). I replaced those drives with new ones. They have been initialized, but when I try to add them to the degraded volume, I get a message saying "Unable to add a disk to this volume." I can't mount the volume either, so what would be the next steps for us to recover the data? All the other drives appear to work well, fyi. I have included the support file. Thank you!
Posted : 03/04/2025 11:33 am
Put the two failing disks back in and see if your volume mounts. Then you can remove ONE disk and "add disk". then let it rebuild. Then you can do this to the second disk. Let us know.
Posted : 03/04/2025 7:04 pm
Topic starter
Thank you! The volume now shows as mounted with the two old failing disks. Should I validate and rebuild it before replacing the first failing disk? The volume does show as degraded, needing rebuilding, and failed.
Posted : 08/04/2025 11:55 am
Is it mounting read/write? Then use "recover failed disks" and validate. Consider updating your backups first.
Then you can replace one, then the other drive.
Posted : 08/04/2025 5:10 pm
Topic starter
I used recover failed disks and now the volume isn't failing. It still says it is degraded and needs rebuilding. One of the failing disks is now showing no errors, which is strange but makes things easier. Any thoughts on why is that happening?
Lastly, would I need to rebuild the volume? I want to make sure I follow all proper steps.
Posted : 09/04/2025 10:24 am
@lilapmr
After "recover failed disks" you need to validate the volume, to regenerate the parity.
Is your volume mounting read only or normal read/write?
Posted : 09/04/2025 5:04 pm
Topic starter
@softraid-support How do I determine if it is mounting read-only or read/write? I tried to use Finder to access the volume, and it crashed.
Posted : 09/04/2025 6:20 pm
MacOS would put up a dialog box telling you.
You can only rebuild a "healthy" volume. Fixiung out of sync, or failed disks have no effect on the contents. The volume either mounts healthy, or not. If the volume is not able to mount properly, the volume cannot rebuild.
But just opening the volume crashed the computer?
Attach a support file, so I can take a look.
Posted : 09/04/2025 10:32 pm
Topic starter
Yes, trying to access the volume on Finder crashed the computer. And the OS has been having issues when restarting, it often freezes completely. And now SoftRaid is not even generating the support file. It loads for a bit after the request and then there is no file anywhere (I made sure to look for it, based on the format of the file I sent before). I will try to restart and see if that helps.
Posted : 10/04/2025 3:10 pm
Topic starter
Here is the file, I had to quit SoftRaid a couple of times for it to produce it.
Posted : 10/04/2025 3:54 pm
Backup IMMEDIATELY. You have two failed disks.
And plan to buy new drives, perhaps a new enclosure, as it appears to be 8 years old. Better to buy the whole bushwack, than just replace the drives.
Do not replace drives, you are just as likely to get another failed disk during the replacement process.
63,000 hours is way past their lifespan, you have been very fortunate so far.
Posted : 10/04/2025 6:58 pm
Topic starter
That's what I said in the beginning of my messages. We do have backups for all these files on cloud storage. I had bought two new drives to replace the failing drives. We may not have the budget right now to buy a new enclosure. Why are you saying that we are likely to get another failed disk during the replacement process? Are you actually suggesting to get all new hard drives and start from scratch with this volume?
Posted : 11/04/2025 9:36 am
@lilapmr
Yes that is exactly what I am saying.
In 30 years, the number of drives I have seen that were healthy at 60,000 hours is a very small percentage.
They are the equivalent of cars with 200,000 miles, but are unrepairable if anything breaks.
If I had to guess, I would say they have a failure probability greater than 25% a year per drive. Yes its possible all or some of the remaining could last another 10,000 hours, but more likely more than one will soon start hanging and causing kernel panics.
In general, the expected life of drives is about 30,000 hours. (they just start failing much more frequently) At 40,000 the failure rates climb quite a bit. Consider yourself fortunate to have all your drives last 60,000.
You can try doing one at a time, but my other point is your enclosure is also at a stage where it can start to fail, with unpredictable behavior.
Posted : 11/04/2025 10:08 am
@lilapmr
I had a suggestion from engineering.
If you end up going through the "replace one drive at a time", and it is successful, send a new support file when done and we will write up a blog post about it!
Posted : 11/04/2025 12:54 pm
Topic starter
@softraid-support
Ok, will do next week, I will update you on it!
Posted : 11/04/2025 4:02 pm
