Conversation
Collaborator
|
Thank you for the contribution. I have issues with your test case: error reports that you are trying to establish a TCP connection, and failure is expected. Let's continue the discussion in https://perconadev.atlassian.net/browse/PT-2510 and get back to this contribution when we are done with the test case. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Ticket number: PT-2485 PT-2510
MariaDB with root socket user. I can connect to the server with
mysqlwithout any password.With percona-toolkit 3.7.0-2.el9 such command works:
With 3.7.1-2.el9 this command does not work:
Root cause:
It's the PT-2305 feature ("pt-online-schema-change should error if server is a slave in row based replication").
In
main(), right after connecting, unless--forceis given, the tool now does:get_source_dsn()runsSHOW REPLICA/SLAVE STATUSand, if your server is itself a replica of some upstream master, builds a DSN pointing at that master's host/port and immediately opens a second connection to it — to check whether replication is safely STATEMENT-based.This breaks two scenarios:
--recursion-method=none: that flag only ever gated downstream replica discovery. This new check ignores it completely, so even though you explicitly said "don't go looking at other servers," it still does.Unix socket auth: your first connection authenticates via the local socket with no
-u/-p(OS-user trust, local only). That trust can't extend to a second, remote TCP connection to the upstream master's hostname — so it tries with no credentials and MySQL/MariaDB correctly reportsAccess denied ... using password: NO. Since$make_cxndies on connection failure, the whole run aborts.In 3.7.0 this second connection simply didn't exist, which is why it worked.
The fix
Skip this check when
--recursion-method=noneis explicitly set (consistent with how the rest of the tool already treats that flag), and make a failed connection to the discovered source non-fatal — warn and skip the safety check instead of killing the whole run, mirroring the existing "don't die on tunnel/port-redirection" handling right below it in the same block.