| Bug #42220 | Agent fails to identify local socket as local on MacOSX 10.4 | ||
|---|---|---|---|
| Submitted: | 20 Jan 2009 16:32 | Modified: | 6 Mar 2009 16:36 |
| Reporter: | Domas Mituzas | Email Updates: | |
| Status: | Closed | Impact on me: | |
| Category: | MySQL Enterprise Monitor: Agent | Severity: | S3 (Non-critical) |
| Version: | 2.0.2.7133 | OS: | MacOS (10.4) |
| Assigned to: | MC Brown | CPU Architecture: | Any |
| Tags: | mem_20_maint | ||
[20 Jan 2009 16:32]
Domas Mituzas
[21 Jan 2009 12:13]
Mark Leith
GDB output looking at agent_mysqld_is_local / network_address_is_local
Attachment: bug42220_gdb.txt (text/plain), 9.69 KiB.
[21 Jan 2009 17:36]
Mark Leith
Test case output: Medusa:~/Desktop mark$ gcc -Wall -O2 unix-getsockname.c -o unix-getsocketname && ./unix-getsocketname our family: 0 their family: 1
[21 Jan 2009 17:36]
Mark Leith
test case
Attachment: unix-getsockname.c (, text), 1.82 KiB.
[28 Jan 2009 20:16]
Mark Leith
In the case where we are connecting with a socket connection, there really is no reason to go down this track anyway - can we not always assume we are local when we are connecting file a socket file *first*?
[2 Mar 2009 22:33]
Jan Kneschke
Patch and test-case have been committed to trunk. revno: 1208 committer: jan@mysql.com branch nick: trunk timestamp: Mon 2009-03-02 23:03:25 +0100 message: fixed network_socket_is_local() for unix-sockets on MacOSX 10.4 (fixes #42220)
[2 Mar 2009 22:53]
Jan Kneschke
Committed and pushed to rel-2.0 (and trunk): revno: 1222 committer: jan@mysql.com branch nick: rel-2.0 timestamp: Mon 2009-03-02 23:50:55 +0100 message: fixed network_socket_connect() to return NETWORK_SOCKET_* as the prototype defines ------------------------------------------------------------ revno: 1221 committer: jan@mysql.com branch nick: rel-2.0 timestamp: Mon 2009-03-02 23:50:38 +0100 message: check that we don't delete a normal file ------------------------------------------------------------ revno: 1220 committer: jan@mysql.com branch nick: rel-2.0 timestamp: Mon 2009-03-02 23:50:20 +0100 message: cleanup socket after we finished the test ------------------------------------------------------------ revno: 1219 committer: jan@mysql.com branch nick: rel-2.0 timestamp: Mon 2009-03-02 23:49:51 +0100 message: added proper checks ------------------------------------------------------------ revno: 1218 committer: jan@mysql.com branch nick: rel-2.0 timestamp: Mon 2009-03-02 23:49:37 +0100 message: fixed network_socket_is_local() for unix-sockets on MacOSX 10.4 (fixes #42220) ------------------------------------------------------------ revno: 1217 committer: jan@mysql.com branch nick: rel-2.0 timestamp: Mon 2009-03-02 23:49:02 +0100 message: unlink() unix-sockets before we bind() to them if no one else is bound to it ------------------------------------------------------------ revno: 1216 committer: jan@mysql.com branch nick: rel-2.0 timestamp: Mon 2009-03-02 23:48:15 +0100 message: moved accept() code down to network_socket_accept() to make the code testable
[5 Mar 2009 16:15]
Diego Medina
Verified fixed on 2.0.5.7150 On the agent logs we see (critical) agent_mysqld.c:643: successfully connected to database at /Users/mysqldev/dv1/data/mysql.sock as user root (with password: YES) (message) agent_mysqld.c:320: [dv1] mysqld is local and directly connected
[6 Mar 2009 16:36]
Tony Bedford
An entry was added to the 2.0.5 changelog: The Agent failed to identify local sockets as local on Mac OS X 10.4. If the Agent was configured to use a Unix domain socket on Mac OS X 10.4, it did not treat the connection as local and failed to provide CPU and memory information to MySQL Enterprise Monitor. This is shown in the log file: 2009-01-20 18:15:02: (message) network-socket.c:752: is-local family 0 != 1 2009-01-20 18:15:02: (message) agent_mysqld.c:322: [mysql] mysqld is not local or not directly connected However, this problem did not happen on Mac OS X 10.5.
