I’ve been running a node for quite a while but lately things haven’t been so smooth. The latest problem I really couldn’t find an answer for on the Internet. Node keeps restarting endlessly and this is the first error I see in the logs:
2020-12-12T14:54:27.997Z ERROR services unexpected shutdown of a runner {“name”: “piecestore:monitor”, “error”: “piecestore monitor: error verifying location and/or readability of storage directory: node ID in file (1M4kqUTsAS8YkE9CkMKrkGtkacJcCraU1ZXgUujuTdh66xjynV) does not match running node’s ID (1gyT7yW5GHiRnsGNEur5E9ATXDNzwrpmT73kQsJW8dVqmNL1nh)”, “errorVerbose”: "piecestore monitor: error verifying location and/or readability of storage directory: node ID in file (1M4kqUTsAS8YkE9CkMKrkGtkacJcCraU1ZXgUujuTdh66xjynV) does not match running node’s ID (1gyT7yW5GHiRnsGNEur5E9ATXDNzwrpmT73kQsJW8dVqmNL1nh)
I previously had an issue with the latest docker version not recognising the directory in which I have my identity stored, so I downgraded to docker 2.5.0.1 and started from scratch creating a new identity, authorizing it and creating a new storagenode docker container.
You managed to mix the identity from the one node with the data from the other node with a different identity. Or you have re-created an identity for the existing node for any reason.
Please, note - the data and identity are tied together, either identity or data is useless without the each other part.
From you description I can say that your first node is disqualified already (but you will not see that, because you do not have that identity anymore), the new identity will not accept the data from a previous one, so the old data will be trashed if you managed to override the check.
If your new identity was not running too long, I would like to suggest to remove the data, the current identity and create a new one, sign it with a new authorization token and setup a new node on clean storage: https://documentation.storj.io/setup/cli/storage-node#setting-up-the-storage-node
Yes, you already lost it with losing its identity.
However, if your new identity is worked for a while, you could have some new data, belongs to that new identity. If it is not running for like a month, I would not bother with overriding the check and start from scratch.
Not only the container, but the data too, unless you have a backup of your previous identity as suggested in the documentation.
If you do not have a backup of your identity, then its data is useless, as you saw already.
I would like to suggest to move your new identity to the disk with data. In such a case they could not be lost independently of each other.
I restored the old identity from the trash and recreated the docker container and now all works well. Thanks a lot Alexey for the quick and effective help!
Hello, so i was also having problems with my node that wouldnt go online and now after 2 days i went to check again and this happened
SO what? If i have any technical issue without even starting my node i have to create a new identity and burn my entire cpu again??? Are you serious?? I have already done this two times but i get errors and now have to create a 3rd identity to finally start the process???
As mentioned in this thread, this really only happens if you mix data from a node with an identity from another. Either use the correct identity or yes, you do have to start over if you don’t have that identity anymore. If it’s really an entirely new node that has never received any data, you can remove everything from the data location and run the setup command again. But if this node has previously received data, it will most likely be disqualified. So starting with a clean identity is your safest bet.
And yes, I am serious, but this can only occur as a result of user error. So cool it on the outrage and we’ll try to help you fix it.
Ok well i checked again and somehow there was data in the mounter directory (even though bandwidth always stayed at N/A.
But thanks anyway i deleted the data and now i am getting other errors
Probably errors about missing data then if that node had received pieces before. If that’s the case you’re really better off starting with a clean identity.
Easy to say but i have already done this twice and my cpu doesn’t seems to like it. If i have to burn a couple of cpu’s before starting then this doesn’t even worth it. The first identity wouldn’t even start the node and now the second just runs but nothing happens i still cannot open my dashboard, although it is running… and the port is still open
