NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Solved
Missing space on RN316 with 5*10TB discs
Hi All,
I have ReadyNas 316 running latest firmware 6.9.3, with 5 times 10T disks (Seagate IronWolf ST10000VN0004) which is on the compatibility list (https://kb.netgear.com/20641/ReadyNAS-Hard...
You had a paused balance. The volume can't be expanded whilst there is a balance in progress. I cancelled the balance and then was immediately able to expand the volume.
Your 4GB root volume is quite full. You should free up some space on it.
Also after I expanded your volume I had a look at smart_history.log and saw that one of your disks is failing.
Done - thanks!
You had a paused balance. The volume can't be expanded whilst there is a balance in progress. I cancelled the balance and then was immediately able to expand the volume.
Your 4GB root volume is quite full. You should free up some space on it.
Also after I expanded your volume I had a look at smart_history.log and saw that one of your disks is failing.
Yout help is very much appreciated!
Few follow-up questions:
- How/why did a balance get "paused"? I am not aware of any way a user can do this?
- When I inserted a new disk, there was nothing on the LED display, or on the web interface, indicating that any maintenance task is running or paused. Maybe this should be indicated somehow to prevent sittuations like the one in OP?
- Cleaning up root volume will be hard to do without removing applications. I did a complete analisys and there is nothing in there that should not be. Is there a way to increase size of root fs?
- On the sidenote, 4G size for root is probably not optimal in year 2018. And so is the fixed size of 1GiB for swap partition (especially since BTRFS does not support swap files)
- Cannot find file named "smart_history.log"?
- No warning was emailed or visible regarding any SMART issues?
- Looking at full SMART reports in Webmin, all I see are 2 drives with few relocated sectors (248 sectors beeing the higest) which I would probably not classify as a "failing" drive... few timeouts too, but those where probably due to the high load at the time or because of relocating sectors. No pending sectors that I can see, so all relocated sectors are well within the reserved relocation space. Did I miss something?
- On a sidenote - with 10TB and higer capacity of drives being comon these days, it is statistically unlikely a drive will go trough it's expected lifespan with zero relocated sectors. IMHO
- How do I find out which drive is having isses? Do you beleive it needs to be replaced? Would you expect a manufacturer to accept that drive as failed?
Thanks you again for yout help!
Andrej
Very nice questions. Just posting to stay in the loop. Thanks
afalout wrote:
Cleaning up root volume will be hard to do without removing applications. I did a complete analisys and there is nothing in there that should not be. Is there a way to increase size of root fs?FWIW, a 4GB OS partition is large enough given the way Netgear intends it to be used. If it is too large, it will impact the data volume space available for some users (particularly users who are deploying SSDs for small/fast volumes or for metadata tiering). So there is a balance here.
In any event, if the OS partition gets completely full, you can corrupt the configuration, so it is best to take care of this.
Apps are normally installed to /data/.apps, which is mounted as /apps in the OS. So they shouldn't take space in the root volume if they are written to Netgear's guidelines. Packages you install with ssh should ideally also be installed onto the data volume.
One thing you could do now is copy some folders they use to /data, and replace the folders with softlinks to the new location.
afalout wrote:
- Looking at full SMART reports in Webmin, all I see are 2 drives with few relocated sectors (248 sectors beeing the higest) which I would probably not classify as a "failing" drive... few timeouts too, but those where probably due to the high load at the time or because of relocating sectors. No pending sectors that I can see, so all relocated sectors are well within the reserved relocation space. Did I miss something?
- On a sidenote - with 10TB and higer capacity of drives being comon these days, it is statistically unlikely a drive will go trough it's expected lifespan with zero relocated sectors. IMHO
There are different views on this I guess. I expect to see no reallocated sectors or pending sectors on a healthy disk, no matter what the size. The largest disks I am using right now 8 TB. My own threshold for replacement is no more than 50 reallocated+pending sectors, and I will replace them before that if the counts are rising. Personally I would replace one that has ~250 reallocated sectors.
I don't think the timeouts can be caused by high load, though it is possible that they are caused by unwriteable sectors.
Have you run the built-in disk test? There might be more bad sectors that you don't know about.
afalout wrote:Would you expect a manufacturer to accept that drive as failed?
I've found that they generally will, though I think that's a secondary question.
afalout wrote:
Cannot find file named "smart_history.log"?
Download the log zip file from the web ui, and look there for it.
1. If a balance was in progress (e.g. an unsafe shutdown occurred during a balance) we wouldn't automatically resume it, but leave it paused.
2. That is a good suggestion.
Yes the disk with 248 reallocated sectors was the one I meant. imo that is quite a lot even for a 10TB disk.
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
