Hi, I commented a similar issue a couple of months ago: https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/message/IQX2VXA6QQQ... Can you check if that applies to your cluster? Zitat von Magnus Larsen <magnusfynbo@hotmail.com>:
Hi Ceph-users!
Ceph version: ceph version 17.2.6 (d7ff0d10654d2280e08f1ab989c7cdf3064446a5) quincy (stable) Using cephadm to orchestrate the Ceph cluster
I’m running into https://tracker.ceph.com/issues/59189, which is fixed in next version—quincy 17.2.7—via https://github.com/ceph/ceph/pull/50906
But I am unable to upgrade to the fixed version because of that bug
When I try to upgrade (using “ceph orch upgrade start –image internal_mirror/ceph:v17.2.7”), we see the same error message: executing _write_files((['dkcphhpcadmin01', 'dkcphhpcmgt028', 'dkcphhpcmgt029', 'dkcphhpcmgt031', 'dkcphhpcosd033', 'dkcphhpcosd034', 'dkcphhpcosd035', 'dkcphhpcosd036', 'dkcphhpcosd037', 'dkcphhpcosd038', 'dkcphhpcosd039', 'dkcphhpcosd040', 'dkcphhpcosd041', 'dkcphhpcosd042', 'dkcphhpcosd043', 'dkcphhpcosd044'],)) failed. Traceback (most recent call last): File "/usr/share/ceph/mgr/cephadm/ssh.py", line 240, in _write_remote_file conn = await self._remote_connection(host, addr) File "/lib/python3.6/site-packages/asyncssh/scp.py", line 922, in scp await source.run(srcpath) File "/lib/python3.6/site-packages/asyncssh/scp.py", line 458, in run self.handle_error(exc) File "/lib/python3.6/site-packages/asyncssh/scp.py", line 307, in handle_error raise exc from None File "/lib/python3.6/site-packages/asyncssh/scp.py", line 456, in run await self._send_files(path, b'') File "/lib/python3.6/site-packages/asyncssh/scp.py", line 438, in _send_files self.handle_error(exc) File "/lib/python3.6/site-packages/asyncssh/scp.py", line 307, in handle_error raise exc from None File "/lib/python3.6/site-packages/asyncssh/scp.py", line 434, in _send_files await self._send_file(srcpath, dstpath, attrs) File "/lib/python3.6/site-packages/asyncssh/scp.py", line 365, in _send_file await self._make_cd_request(b'C', attrs, size, srcpath) File "/lib/python3.6/site-packages/asyncssh/scp.py", line 343, in _make_cd_request self._fs.basename(path)) File "/lib/python3.6/site-packages/asyncssh/scp.py", line 224, in make_request raise exc asyncssh.sftp.SFTPFailure: scp: /tmp/etc/ceph/ceph.conf.new: Permission denied During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/share/ceph/mgr/cephadm/utils.py", line 79, in do_work return f(*arg) File "/usr/share/ceph/mgr/cephadm/serve.py", line 1088, in _write_files self._write_client_files(client_files, host) File "/usr/share/ceph/mgr/cephadm/serve.py", line 1107, in _write_client_files self.mgr.ssh.write_remote_file(host, path, content, mode, uid, gid) File "/usr/share/ceph/mgr/cephadm/ssh.py", line 261, in write_remote_file self.mgr.wait_async(self._write_remote_file( File "/usr/share/ceph/mgr/cephadm/module.py", line 615, in wait_async return self.event_loop.get_result(coro) File "/usr/share/ceph/mgr/cephadm/ssh.py", line 56, in get_result return asyncio.run_coroutine_threadsafe(coro, self._loop).result() File "/lib64/python3.6/concurrent/futures/_base.py", line 432, in result return self.__get_result() File "/lib64/python3.6/concurrent/futures/_base.py", line 384, in __get_result raise self._exception File "/usr/share/ceph/mgr/cephadm/ssh.py", line 249, in _write_remote_file logger.exception(msg) orchestrator._interface.OrchestratorError: Unable to write dkcphhpcmgt028:/etc/ceph/ceph.conf: scp: /tmp/etc/ceph/ceph.conf.new: Permission denied
We were thinking about removing the keyring from the Ceph orchestrator (https://docs.ceph.com/en/latest/cephadm/operations/#putting-a-keyring-under-...), which would then make Ceph not try to copy over a new ceph.conf, alleviating the problem (https://docs.ceph.com/en/latest/cephadm/operations/#client-keyrings-and-conf...), but in doing so, Ceph will kindly remove the key from all nodes (https://docs.ceph.com/en/latest/cephadm/operations/#disabling-management-of-...) leaving us without the admin keyring. So that doesn’t sound like a path we want to take :S
Does anybody know how to get around this issue, so I can get to version where the issue fixed for good?
Thanks, Magnus _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io