Jaster Posted August 16, 2020 Share Posted August 16, 2020 (edited) I had a dropped disk (disk1) and found out it had tons of CRC errors. So I changed the cables. Here I attached the recent diagnostics - what do you think? Keep it or replace it? knowlage-diagnostics-20200816-1418.zip P.S: I'm trying to run a simple smart test, but it seems to hang at 90%... doesn't look good?... Edited August 16, 2020 by Jaster Quote Link to comment
Frank1940 Posted August 16, 2020 Share Posted August 16, 2020 CRC error count is not resetable. Once a disk has had a CRC error, it will follow the disk forever. Basically what this means is that the only important thing about CRC errors is to determine if they are still increasing. (CRC errors are data transmission errors and they are corrected by transmitting the data. However, they do slow data transfer. Plus, some folks feel that if a very high rate of these errors could allow could result in a packet of data being accepted because the errors compute in such a way to produce a correct CRC result.) About the failed SMART test. That is a concern. Plus there are errors in the SMART log. I am no expert in this area. I always have a spare disk ready to go for such occurrences. I replace the questionable disk and then run a series of preclear cycles on it to determine what to do next with it. Quote Link to comment
Jaster Posted August 16, 2020 Author Share Posted August 16, 2020 So basically a 'maybe'. Time to replace it then. Not the first time that guy dropped. If someone with more understanding of smart reads this, I'd welcome more input. Meanwhile waiting for that spare to arrive Quote Link to comment
Frank1940 Posted August 16, 2020 Share Posted August 16, 2020 Let's see if Johnnie Black has the time to have a look at this. @johnnie.black He has the expertise to interrupt the contents of SMART logs. Quote Link to comment
Michael_P Posted August 16, 2020 Share Posted August 16, 2020 The CRC errors for disk 1 don't look bad, it's the reallocated sectors for your parity disk VKK45Z3Y that would concern me Quote Link to comment
Frank1940 Posted August 16, 2020 Share Posted August 16, 2020 Comments on the Reallocated sectors on the parity disk. 1-- The Reallocated Sector count is in the single digits. 2-- There are no Pending Sectors This means that all of the sectors currently in use have not have a problem detected on them. Now, the HD manufacturers provide a pool of spare sectors to be used when one or more in-use sectors have problems. (Each manufacturer has a slightly different scheme for how it defines, detects and handles problematic sectors.) When a problematic sector is detected, it marked as 'Pending'. As I understand it, the sector will be handled when the next write operation occurs. In some cases, it might be 'cleared' or, in many cases, a new sector will simply be 'swapped' in for the old one. With the counts being as they are in this case, many experts recommend to continue to use the disk BUT keep an eye on it. IF one tends toward paranoia, one might want to replace it and stress test it by running a series of preclear operations on it to see if the disk might be in the very early stage of failure. OP, may we assume that you are running periodical parity checks on your array and you have the notification system setup to provide reports of any issues? Quote Link to comment
JorgeB Posted August 17, 2020 Share Posted August 17, 2020 18 hours ago, Jaster said: Keep it or replace it? Disk look perfectly fine, though wait for the extended test to finish to confirm. Quote Link to comment
Jaster Posted August 17, 2020 Author Share Posted August 17, 2020 Not sure what's happening. Disks keeps dropping and today the suspicious parity disk dropped. @johnnie.black I had a hard time finishing even the short test, but I'll try and trigger the extended. Waiting for a spare to arrive and start replacing disks just to verify if they are good. Quote Link to comment
testdasi Posted August 17, 2020 Share Posted August 17, 2020 53 minutes ago, Jaster said: Not sure what's happening. Disks keeps dropping and today the suspicious parity disk dropped. @johnnie.black I had a hard time finishing even the short test, but I'll try and trigger the extended. Waiting for a spare to arrive and start replacing disks just to verify if they are good. How are you powering the drives? Poor quality power cable / overloading too many drives on the same power cable / too long power cable can cause drives to drop offline randomly. Quote Link to comment
Jaster Posted August 17, 2020 Author Share Posted August 17, 2020 My PSU is quite overpowered (corsair 1200) and I seem not to have ran into those kind of issues yet. I'm trying to use mostly the PSU power cables, some adapters are needed. Worst case would be around 8 drives at one PSU rail. Quote Link to comment
Michael_P Posted August 17, 2020 Share Posted August 17, 2020 How many on 1 cable back to the PSU? Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.