aarch64 is the response from uname
If that’s ok, the Time Machine idea seems to be the most logical one so far.
I still have these in the second, new node (only this and the 0 online score for just one satellite):
2022-02-01T18:05:43.056Z ERROR orders cleaning filestore archive {"error": "order: lstat config/orders/archive/archived-orders-1wFTAgs9DP5RSnCqKV1eLf6N9wtk4EAtmN5DpSxcs8EjT69tGE-1643569200000000000-1643576684730631097-ACCEPTED.v1: input/output error", "errorVerbose": "order: lstat config/orders/archive/archived-orders-1wFTAgs9DP5RSnCqKV1eLf6N9wtk4EAtmN5DpSxcs8EjT69tGE-1643569200000000000-1643576684730631097-ACCEPTED.v1: input/output error\n\tstorj.io/storj/storagenode/orders.(*FileStore).CleanArchive.func1:366\n\tpath/filepath.walk:438\n\tpath/filepath.Walk:505\n\tstorj.io/storj/storagenode/orders.(*FileStore).CleanArchive:364\n\tstorj.io/storj/storagenode/orders.(*Service).CleanArchive:163\n\tstorj.io/storj/storagenode/orders.(*Service).Run.func2:141\n\tstorj.io/common/sync2.(*Cycle).Run:152\n\tstorj.io/common/sync2.(*Cycle).Start.func1:71\n\tgolang.org/x/sync/errgroup.(*Group).Go.func1:57"}
HDD checked, directories readable and writable. Besides the single satellite, everything else is at 100% on node no. 2.
Update: another option is, the icy box 4 bay is not as reliable as required, causing I/o issues. Will try to get it replaced, while node1 is currently suspended and node2 offline.
Iwill move it back to RPi and connect it to there directly. While moving, I need to format the hdd with ext4 (right?).
Should i start rsync while the node is running or should I take it offline? At least for the last rsync, I know - but for the first sync, too?
No, the point of multiple rsync runs is that you can keep it running while copying the bulk and then only have to copy what changed. You CAN stop it right away, but there’s not really a need to use rsync then. A regular copy would do just fine. But you’ll have longer downtime.
That’s what I wanted to know, thank you.
Hey there @Stob @BrightSilence @Alexey - just wanted to give an update here:
I’ve moved node#1 back to my RPi, directly connected to the IcyBox. Since then, it is running for more than 140h without ANY issues, all success rates are at (almost) 100%, constantly. That’s really a good feeling, and I can leave the system (again) untouched and running.
Thank you again for your reliable help!
4 Likes