rendermaster
-
Posts
9 -
Joined
-
Last visited
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Posts posted by rendermaster
-
-
welp. after a reboot and restarting the array. it appears to be good. recent diagnostics attached.
guess i panicked a bit and didn't think to just restart the system
-
I have 2 sata SSDs in a pool "VM" for docker, appdata, VMs etc. this morning it is showing as "unmountable: no file system" I'm searching forums but wanted to post diagnostics here before I take any action.
This data isn't un-replacable but it would be a major pain to lose the docker images and app data. but i think I have the data also on my main spinning array..?
-
I'm using the icloudPD docker container to download my iCloud photo library. It's working but I am getting a lot of IOErrors. I've attached a sample below. I had thought it was maybe HDD speed but I turned on cache on a SSD and I'm still getting the errors, it doesn't appear they are related to file size either, it will download large videos fine, then fail on a basic .jpg or .heic image.
Quote023-01-23 19:32:30 ERROR IOError while writing file to /home/user/iCloud/2020/09/03/IMG_5040.HEIC! You might have run out of disk space, or the file might be too large for your OS. Skipping this file...
2023-01-23 19:32:30 INFO Downloading /home/user/iCloud/2020/09/03/IMG_5039.HEIC
2023-01-23 19:32:31 INFO Downloading /home/user/iCloud/2020/09/03/IMG_5038.HEIC
2023-01-23 19:32:31 INFO Downloading /home/user/iCloud/2020/09/03/IMG_5037.HEIC
2023-01-23 19:32:31 ERROR IOError while writing file to /home/user/iCloud/2020/09/03/IMG_5037.HEIC! You might have run out of disk space, or the file might be too large for your OS. Skipping this file...
2023-01-23 19:32:31 INFO Downloading /home/user/iCloud/2020/09/03/IMG_5036.HEIC
2023-01-23 19:32:32 ERROR IOError while writing file to /home/user/iCloud/2020/09/03/IMG_5036.HEIC! You might have run out of disk space, or the file might be too large for your OS. Skipping this file... -
It was because I was using Safari (better battery life on M1 hardware). Chrome and Firefox both give a normal zip file. Latest diagnostics attached. But I think I got everything sorted. Got docker and VMs setup on a pool of 2 120GB ssds and moved the cache to the 1TB m.2. It's all green lights and the fix common problems plugin also looks happy.
Thanks again for your quick response, kindness and help with this @trurl I'm looking forward to digging into unraid more; Home Assistant, network ad blocking and DNS filtering, windows VM for some older games, offsite backups and some personal cloud stuff.
-
2 hours ago, trurl said:
Many more common ways to lose data, including user error.
Why? Diagnostics are downloaded as a single zip file. If it appears to you that it is downloaded as multiple folders and files, it is because your browser or computer is automatically opening zip files. You should be able to find the original zip wherever your downloads go.
user error. hadn't really thought of that, but of course, with full read/write permissions to a share a user could just start deleting data
diagnostics file was because I use MacOS and safari on my laptop and wanted to get it done quickly. Was the file still functional?
-
2 hours ago, trurl said:
cache:prefer means prefer to keep this share on cache, and if anything overflows to the array, move to cache if it can.
^^^ This is the main reason you are filling cache. You have all your shares set to prefer cache.
To get a share moved from cache to array, it must be set to cache:yes.
THANK YOU. and thank you for answering with kindness when this is very much a RTM situation. I very much appreciate you help here. I was able to reassign the m.2 drive to the cache and the errors immediately cleared. and I'll put in that second SATA ssd after work today and build a new pool for docker/VMs.
-
9 hours ago, trurl said:
That is going to be too small for caching user share writes. Might be enough for docker/VM shares appdata, domains, system.
Any files on single disk pool such as your cache are unprotected. Not necessarily a bad thing, just something to be aware of. You will want backups of anything important and irreplaceable anyway. Parity is not a substitute.
Ahhh.... I see. Is there a recommend size for cache drive? I also have a 1TB m.2 drive in the system and a second 120GB sata drive available, although it's a different brand and age of the other 120GB. What is the best way to get VMs/Docker onto the flash storage? I only have one docker container running right now and I've noticed it doesn't save preferences, each time I launch the app it's like the first time it's run. I would like to use a few more docker apps and a VM at some point.
re: "Parity is not a substitute." Parity protects from a single disk failure, correct? I also understand that my server would be vulnerable to physical/geographic issues, fire, flood, theft and plan to eventually back up the entire server to Backblaze/glacier/something. Doesn't the mover service copy the data from all pool devices to the array nightly? Thereby protecting from a single (pool or array) disk failure?
Diagnostics attached. I had to re-zip it after the download. If that causes a problem I can post again.
-
I have 4x4TB drives (1 parity, 3 data) as my main storage array with a 120GB sata SSD for cache. I followed the basic setup instructions but even after a few days the cache drive is always completely full and my shares appdata, domains and system show as "unprotected". I've RTM a bit but I just don't see what I'm missing.
(PS-why is blanking out the serial numbers of a disk something? I don't immediately see how that data could be a security issue, could someone enlighten me?)
IOError with icloudPD
in Docker Engine
Posted
It seems to have resolved itself after running for a few days. I'm getting working syncs now with no errors.