This topic has been marked solved and closed to new posts due to inactivity. We hope you'll join the conversation by posting to an open topic or starting a new one.
After creating a new share on my RN202 I went on and created a rsync backup job on the Duo.
I saw that snapshots needed to be excluded, so cancelled the backup job. The Duo advised that I should resynch the disks, so it was started.
It would take a long time so I was not concerned that the Duo was not responding to anything, except I could open the first level menus on the Admin page. The next day was status quo, including my connection to the Admin page was still active. However, closing the window and reopen the connection does not show the first level menus.
Status of the Duo is visible in RAIDar-4.3.8. First column led and both disk leds are green; but the led just after the IP-address is blinking green/yellow and the mouse displays: Volume C: Raid level X, Resync 0% complete, Time to finish 13 hr 30 min, Speed 40MB/s; 1115 GB (60%) of 1853 GB used This status has not changed for at least 24 hours.
The disk leds on the Duo seems to be off.
For what it's worth, I have a .ZIP file with logs dated 20230326, taken after I managed to install the fix "Workaround for ERR_SSL_VERSION_OR_CIPHER_MISMATCH problem" so SSH might still be installed. Thank you for the step by step instructions.
The Duo is a backup unit, my primary server is OK, and I have an extra disk backup.
Is it advisable to power off and hope the server will be OK after reboot, or can I do anything else to fix the resync?
Putty could not connect to my Duo, so I figured that next step should be testing the disks using WD diagnostics tool. I have used it before so no problem, but it does take some hours per disk. Follow-up tomorrow.
I do use the older SW and have not found problems with old disks.
And you also used the full write-zeros test in Lifeguard?
Also what disks are you using? Many new drives 2 TB and under are SMR. They can have very slow sustained write times, so operations like resync could take quite a while.
I used the FULL erase function, but I don't know the kind of data it writes.
My disks are WD20EFRX and I have spend some time educating myself about the difference between WD20EFAX and WD20EFZX, and then what is CMR vs SMR. I suppose the compatibility-list is no longer updated, so I have chosen to stay with the CMR version.
I suppose the compatibility-list is no longer updated,
Very true. I always recommend NAS-purposed CMR drives for the Duo and NV+ models. WD Red Plus (not Reds) and Seagate IronWolf are good choices. Enterprise-class drives are also fine, but IMO overkill, given the limited performance of these particular models.
Note that the WD20EFRX is CMR (and is now branded as a WD Red Plus).
PMR is an older name for CMR (for a specific CMR technology).
The general issue with SMR is that writing to track X destroys the data on track X+1. That creates a rippling
read track X+1, write track X,
read track X+2, write track X+1
read track X+3, write track X+2
...
That continues until you reach the end of the SMR zone (where there is an unused track). The manufacturers overcome the obvious write performance issue with a lot of cache and attention to the caching algorithm. But sustained writes will overhelm that eventually, and you will always end up with glacially slow write performance at some point.
Best not to use these drives with RAID in any form. Both WD and Seagate are now identifying SMR/CMR on their datasheets.
