NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
ReadyNAS Duo v1 odd behaviour
After the events last week, I thought stability had returned but it seems not....
Last week involved a failing>failed disk. I replaced it with new like for like, resync completed and all was normal...
Today's update:
The NAS had been in resync overnight, recalling an estimate of approx 12hrs to complete, and I was ok with that. When the chance arose I checked where it was at using the NAS Utils iPad app, and could see estimated time reducing, expecting it to complete approx 0800hrs here.
As that time came and went, no emails were received and on checking once more via that app it was still showing resync with an estimated time of 333+ hours! Trying not to be impatient (!) I have left it doing its thing and noted that time estimate reducing quickly.
As I write this it's estimating 66 hours. The time estimates don't seem to be linear so I really don't know how meaningful this is, noting the estimate doesn't seem to be reducing as quickly now.
I don't understand the text next to RAID Status, currently reading "Resync 14&37...etc" I've noticed the first figure is incrementing but not the second.
Other notes:
-Using a file browser app, shares and content seem unaffected currently
-Using RAIDar both drives show as healthy (see pic) but sync is showing 0%. Now it did this before despite a successful sync, I never saw that figure change, so perhaps the app has a bug
-SMART logs show nothing that grabs my attention in terms of ATA errors or uncorrectables/reallocations. The only thing there is Lp stat events for Disk1 (the original of the pair) which is currently at 35831, compared to about 7 for Disk2 (the new disk)
Whilst I have my doubts, this could all be perfectly normal as it sorts itself out, but being unable to find any info on a few points leaves me seeking help here:
-Why it did this without being asked to, and without any warnings or emails. Main log merely said the volume scan found/corrected errors, then RAID sync started
-No indicators on the front panel or within RAIDar etc to show a disk is in a failing state
-Why the sync that started last night seemed to reach its estimated time, then next look showed that huge estimate
-What those figures in the NAS Utils app mean (now showing "Resync 15&37..Time to finish 60hrs etc"
I'd welcome any thoughts at all on this, thanks in advance.
The rapidly rising lp stat counts could indicate that the Duo is having some issues with the first disk. "lp stat" is monitored by the Duo itself - it's not a SMART stat, and I've never seen a clear explanation of what it is.
It would be wise to update your backup.
Clearly you are seeing a formatting bug on the resync page (looks like RAIDar?). As far as the times go, the estimates are often way off, especially on the older legacy NAS.
Thanks for your reply StephenB. It's now down to just over 11 hours pending. If it gets to the end of this sync without a total failure, I'll reboot, check logs and content accessibility - and if all is well, replace Disk1 also (which I know will kick off another resync, but that didn't take too long last week) I hope it doesn't get to the end of this resync and then immediately kick off again, as that seems to be what happened early this morning. Weird thing is that since I replaced Disk2 last week, it's been running perfectly normally, not a hint of more troubles until last night.
That info from NAS Utils (RAID Status) which currently shows "Resync 55%complete, Time to finish etc" interests me. I am wondering if the 55 (which seems to be incrementing) relates to Disk1 and 37 (static) relates to Disk2 perhaps, and if that is some indication of Disk1 being problematic. I'm perhaps clutching at straws given the lack of other clues.
What I don't get is that all other aspects of health that I can see suggest all is good. The only clue is the Lp stat count. As you can see from my pics, everything else shows as green. I'd almost be happier if Disk1 showed a yellow on Frontview, which would at least reinforce something not being right.
Based on another thread here, the suggestion is that Lp stat increase means I/O requests are taking longer than expected. I guess that would stack up if there was a clear fault developing, but as I said all the other indicators such as reallocation, uncorrectables and ATA errors show nothing bad in the SMART summary.
Out of interest do you know of any specific log area I could check to see if there is something fundamentally amiss with the NAS itself that could cause this? Paranoia is setting in here. I can understand the possibility of Disk1 following in the footsteps of Disk2 as they were the same age, and perhaps the recent hammering has pushed it to the edge, I don't know.
Numb3r6 wrote:
That info from NAS Utils (RAID Status) which currently shows "Resync 55%complete, Time to finish etc" interests me. I am wondering if the 55 (which seems to be incrementing) relates to Disk1 and 37 (static) relates to Disk2 perhaps, and if that is some indication of Disk1 being problematic. I'm perhaps clutching at straws given the lack of other clues.
No. % is the html code for %. So it's a minor formatting bug, meaning to say "Resync 55% complete"
Numb3r6 wrote:
Out of interest do you know of any specific log area I could check to see if there is something fundamentally amiss with the NAS itself that could cause this?
Maybe look in system.log and kernel.log for disk errors.
Overall, it is an old NAS. You should plan on replacing it at some point, perhaps with an RN212 or an RN422. Then repurpose the Duo as a backup NAS.
Thanks StephenB, the resync finally finished yesterday evening. I recall from ages ago that time estimates can be wildly out, as you said earlier. I'm now assuming that the sync that started just after 2200 on Thursday, which informed of expected completion by about 0800 Friday, was just not accurate, and the resync actually finished last night. Still no idea what prompted it as I can find nothing untoward in the SMART info or logs at all, bar what looks like a minor glitch with timeserver sync.
I may well have a look at the chassis models you suggest. Would I be right in assuming if I bought diskless I could start off just by inserting a current healthy drive from my Duo, make sure that is ok, then add a second new drive to sync up - or do the newer chassis have a different architecture requiring blank drives and data population by loading up from another source perhaps? Right now there's no pressing need to go beyond 2TB and I don't want to incur too much in the way of extra cost just now.
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
