Re: signal handling in mysql cli
Hello Sebastien, You should stop top-posting, it inverts the flow of your investigation making it harder to follow. (see bottom) On 6/21/2016 12:45 PM, Sebastien FLAESCH wrote: The process list show my (killed) thread as follows: mysql> show processlist; ++-+-+---+-+--+--+--+ | Id | User| Host| db| Command | Time | State| Info | ++-+-+---+-+--+--+--+ | 20 | root| localhost:48203 | test1 | Query |0 | starting | show processlist | | 21 | mysuser | localhost:48209 | test1 | Sleep | 182 | | NULL | ++-+-+---+-+--+--+--+ Why it is in "Sleep" state?!? Seb On 06/21/2016 06:27 PM, Sebastien FLAESCH wrote: FYI, I get the same problem with MySQL 5.7.13. Seb On 06/21/2016 04:59 PM, Sebastien FLAESCH wrote: Seems that after KILL QUERY mysql-thread-id, a call to mysql_stmt_close(stmt-handle) hangs... This did not happen in 5.6 ... Will try 5.7.13 ... Seb On 06/21/2016 04:03 PM, Sebastien FLAESCH wrote: Hi all, The technique described in this thread is working fine with MySQL 5.6 (libmysqlclient). Basically, in a SIGINT signal handler, we establish a new connection to perform a KILL QUERY pid ... But with 5.7 (5.7.11) we get now a different result: A) The query is still interrupted, but we no longer get an SQL error -1317. B) For some reason, the program does not want to exit() - (must investigate) Any clues? With mysql it's working fine: mysql> select sleep(10); ^C^C -- query aborted +---+ | sleep(10) | +---+ +---+ 1 row in set (2.79 sec) mysql> \q Bye Thanks! Seb On 12/03/2014 05:25 PM, Sebastien FLAESCH wrote: Hi all, I have a similar question regarding KILL QUERY usage: We have a C client program using libmysqlclient.so, it is a single-threaded program. When running a long query, how can I send the KILL QUERY command when a SIGINT (CTRL-C) is caught? (of course we implement a signal handler, so we keep the control) => Is is safe to establish a new connection to the server in the signal handler, using mysql_init() + mysql_real_connect(), and execute the KILL QUERY with the mysql thread id I got from the initial mysql_init() / mysql_real_connect()? I made some tests, and it seems to work fine, the long query returns SQL error -1317: "Query execution was interrupted" (which is expected) We want to support SQL interruption properly, so please someone from the dev team, give me a clear answer... I don't want to use a side effect or undocumented feature. Doing all this stuff in a signal handler is certainly risky... no? I could not find the information in the documentation (mysql_real_connect). I wish there would be an API like mysql_cancel_query(), similar to Oracle's OCI OCIBreak(). Thanks! Seb On 12/02/2014 05:13 PM, walter harms wrote: hi list, when i use CTRL-C to break a query that works fine in interactive mode. mysql> select sleep(10) ; ^CCtrl-C -- sending "KILL QUERY 24289" to server ... Ctrl-C -- query aborted. +---+ | sleep(10) | +---+ +---+ 1 row in set (0.86 sec) but when i use the noninteractive mode timeout 5 mysql -BAN -e "select now(); select sleep (100) ; select now() " i looks like that but "show full processlist;" shows otherwise and that is true as a list of long running querys showed. Is there a way to make it behave like the interactive version ? Now it is a bit confusing for everyone. re, wh You did not KILL the entire client session(CONNECTION), you only killed the QUERY that the session was executing. The client remains connected and the session remains active. The reason it is in Sleep state is because the server is waiting for the client to send its next command. http://dev.mysql.com/doc/refman/5.7/en/kill.html -- Shawn Green MySQL Senior Principal Technical Support Engineer Oracle USA, Inc. - Integrated Cloud Applications & Platform Services Office: Blountville, TN Become certified in MySQL! Visit https://www.mysql.com/certification/ for details. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
The process list show my (killed) thread as follows: mysql> show processlist; ++-+-+---+-+--+--+--+ | Id | User| Host| db| Command | Time | State| Info | ++-+-+---+-+--+--+--+ | 20 | root| localhost:48203 | test1 | Query |0 | starting | show processlist | | 21 | mysuser | localhost:48209 | test1 | Sleep | 182 | | NULL | ++-+-+---+-+--+--+--+ Why it is in "Sleep" state?!? Seb On 06/21/2016 06:27 PM, Sebastien FLAESCH wrote: FYI, I get the same problem with MySQL 5.7.13. Seb On 06/21/2016 04:59 PM, Sebastien FLAESCH wrote: Seems that after KILL QUERY mysql-thread-id, a call to mysql_stmt_close(stmt-handle) hangs... This did not happen in 5.6 ... Will try 5.7.13 ... Seb On 06/21/2016 04:03 PM, Sebastien FLAESCH wrote: Hi all, The technique described in this thread is working fine with MySQL 5.6 (libmysqlclient). Basically, in a SIGINT signal handler, we establish a new connection to perform a KILL QUERY pid ... But with 5.7 (5.7.11) we get now a different result: A) The query is still interrupted, but we no longer get an SQL error -1317. B) For some reason, the program does not want to exit() - (must investigate) Any clues? With mysql it's working fine: mysql> select sleep(10); ^C^C -- query aborted +---+ | sleep(10) | +---+ +---+ 1 row in set (2.79 sec) mysql> \q Bye Thanks! Seb On 12/03/2014 05:25 PM, Sebastien FLAESCH wrote: Hi all, I have a similar question regarding KILL QUERY usage: We have a C client program using libmysqlclient.so, it is a single-threaded program. When running a long query, how can I send the KILL QUERY command when a SIGINT (CTRL-C) is caught? (of course we implement a signal handler, so we keep the control) => Is is safe to establish a new connection to the server in the signal handler, using mysql_init() + mysql_real_connect(), and execute the KILL QUERY with the mysql thread id I got from the initial mysql_init() / mysql_real_connect()? I made some tests, and it seems to work fine, the long query returns SQL error -1317: "Query execution was interrupted" (which is expected) We want to support SQL interruption properly, so please someone from the dev team, give me a clear answer... I don't want to use a side effect or undocumented feature. Doing all this stuff in a signal handler is certainly risky... no? I could not find the information in the documentation (mysql_real_connect). I wish there would be an API like mysql_cancel_query(), similar to Oracle's OCI OCIBreak(). Thanks! Seb On 12/02/2014 05:13 PM, walter harms wrote: hi list, when i use CTRL-C to break a query that works fine in interactive mode. mysql> select sleep(10) ; ^CCtrl-C -- sending "KILL QUERY 24289" to server ... Ctrl-C -- query aborted. +---+ | sleep(10) | +---+ +---+ 1 row in set (0.86 sec) but when i use the noninteractive mode timeout 5 mysql -BAN -e "select now(); select sleep (100) ; select now() " i looks like that but "show full processlist;" shows otherwise and that is true as a list of long running querys showed. Is there a way to make it behave like the interactive version ? Now it is a bit confusing for everyone. re, wh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
FYI, I get the same problem with MySQL 5.7.13. Seb On 06/21/2016 04:59 PM, Sebastien FLAESCH wrote: Seems that after KILL QUERY mysql-thread-id, a call to mysql_stmt_close(stmt-handle) hangs... This did not happen in 5.6 ... Will try 5.7.13 ... Seb On 06/21/2016 04:03 PM, Sebastien FLAESCH wrote: Hi all, The technique described in this thread is working fine with MySQL 5.6 (libmysqlclient). Basically, in a SIGINT signal handler, we establish a new connection to perform a KILL QUERY pid ... But with 5.7 (5.7.11) we get now a different result: A) The query is still interrupted, but we no longer get an SQL error -1317. B) For some reason, the program does not want to exit() - (must investigate) Any clues? With mysql it's working fine: mysql> select sleep(10); ^C^C -- query aborted +---+ | sleep(10) | +---+ +---+ 1 row in set (2.79 sec) mysql> \q Bye Thanks! Seb On 12/03/2014 05:25 PM, Sebastien FLAESCH wrote: Hi all, I have a similar question regarding KILL QUERY usage: We have a C client program using libmysqlclient.so, it is a single-threaded program. When running a long query, how can I send the KILL QUERY command when a SIGINT (CTRL-C) is caught? (of course we implement a signal handler, so we keep the control) => Is is safe to establish a new connection to the server in the signal handler, using mysql_init() + mysql_real_connect(), and execute the KILL QUERY with the mysql thread id I got from the initial mysql_init() / mysql_real_connect()? I made some tests, and it seems to work fine, the long query returns SQL error -1317: "Query execution was interrupted" (which is expected) We want to support SQL interruption properly, so please someone from the dev team, give me a clear answer... I don't want to use a side effect or undocumented feature. Doing all this stuff in a signal handler is certainly risky... no? I could not find the information in the documentation (mysql_real_connect). I wish there would be an API like mysql_cancel_query(), similar to Oracle's OCI OCIBreak(). Thanks! Seb On 12/02/2014 05:13 PM, walter harms wrote: hi list, when i use CTRL-C to break a query that works fine in interactive mode. mysql> select sleep(10) ; ^CCtrl-C -- sending "KILL QUERY 24289" to server ... Ctrl-C -- query aborted. +---+ | sleep(10) | +---+ +---+ 1 row in set (0.86 sec) but when i use the noninteractive mode timeout 5 mysql -BAN -e "select now(); select sleep (100) ; select now() " i looks like that but "show full processlist;" shows otherwise and that is true as a list of long running querys showed. Is there a way to make it behave like the interactive version ? Now it is a bit confusing for everyone. re, wh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
Seems that after KILL QUERY mysql-thread-id, a call to mysql_stmt_close(stmt-handle) hangs... This did not happen in 5.6 ... Will try 5.7.13 ... Seb On 06/21/2016 04:03 PM, Sebastien FLAESCH wrote: Hi all, The technique described in this thread is working fine with MySQL 5.6 (libmysqlclient). Basically, in a SIGINT signal handler, we establish a new connection to perform a KILL QUERY pid ... But with 5.7 (5.7.11) we get now a different result: A) The query is still interrupted, but we no longer get an SQL error -1317. B) For some reason, the program does not want to exit() - (must investigate) Any clues? With mysql it's working fine: mysql> select sleep(10); ^C^C -- query aborted +---+ | sleep(10) | +---+ +---+ 1 row in set (2.79 sec) mysql> \q Bye Thanks! Seb On 12/03/2014 05:25 PM, Sebastien FLAESCH wrote: Hi all, I have a similar question regarding KILL QUERY usage: We have a C client program using libmysqlclient.so, it is a single-threaded program. When running a long query, how can I send the KILL QUERY command when a SIGINT (CTRL-C) is caught? (of course we implement a signal handler, so we keep the control) => Is is safe to establish a new connection to the server in the signal handler, using mysql_init() + mysql_real_connect(), and execute the KILL QUERY with the mysql thread id I got from the initial mysql_init() / mysql_real_connect()? I made some tests, and it seems to work fine, the long query returns SQL error -1317: "Query execution was interrupted" (which is expected) We want to support SQL interruption properly, so please someone from the dev team, give me a clear answer... I don't want to use a side effect or undocumented feature. Doing all this stuff in a signal handler is certainly risky... no? I could not find the information in the documentation (mysql_real_connect). I wish there would be an API like mysql_cancel_query(), similar to Oracle's OCI OCIBreak(). Thanks! Seb On 12/02/2014 05:13 PM, walter harms wrote: hi list, when i use CTRL-C to break a query that works fine in interactive mode. mysql> select sleep(10) ; ^CCtrl-C -- sending "KILL QUERY 24289" to server ... Ctrl-C -- query aborted. +---+ | sleep(10) | +---+ +---+ 1 row in set (0.86 sec) but when i use the noninteractive mode timeout 5 mysql -BAN -e "select now(); select sleep (100) ; select now() " i looks like that but "show full processlist;" shows otherwise and that is true as a list of long running querys showed. Is there a way to make it behave like the interactive version ? Now it is a bit confusing for everyone. re, wh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
Hi all, The technique described in this thread is working fine with MySQL 5.6 (libmysqlclient). Basically, in a SIGINT signal handler, we establish a new connection to perform a KILL QUERY pid ... But with 5.7 (5.7.11) we get now a different result: A) The query is still interrupted, but we no longer get an SQL error -1317. B) For some reason, the program does not want to exit() - (must investigate) Any clues? With mysql it's working fine: mysql> select sleep(10); ^C^C -- query aborted +---+ | sleep(10) | +---+ +---+ 1 row in set (2.79 sec) mysql> \q Bye Thanks! Seb On 12/03/2014 05:25 PM, Sebastien FLAESCH wrote: Hi all, I have a similar question regarding KILL QUERY usage: We have a C client program using libmysqlclient.so, it is a single-threaded program. When running a long query, how can I send the KILL QUERY command when a SIGINT (CTRL-C) is caught? (of course we implement a signal handler, so we keep the control) => Is is safe to establish a new connection to the server in the signal handler, using mysql_init() + mysql_real_connect(), and execute the KILL QUERY with the mysql thread id I got from the initial mysql_init() / mysql_real_connect()? I made some tests, and it seems to work fine, the long query returns SQL error -1317: "Query execution was interrupted" (which is expected) We want to support SQL interruption properly, so please someone from the dev team, give me a clear answer... I don't want to use a side effect or undocumented feature. Doing all this stuff in a signal handler is certainly risky... no? I could not find the information in the documentation (mysql_real_connect). I wish there would be an API like mysql_cancel_query(), similar to Oracle's OCI OCIBreak(). Thanks! Seb On 12/02/2014 05:13 PM, walter harms wrote: hi list, when i use CTRL-C to break a query that works fine in interactive mode. mysql> select sleep(10) ; ^CCtrl-C -- sending "KILL QUERY 24289" to server ... Ctrl-C -- query aborted. +---+ | sleep(10) | +---+ +---+ 1 row in set (0.86 sec) but when i use the noninteractive mode timeout 5 mysql -BAN -e "select now(); select sleep (100) ; select now() " i looks like that but "show full processlist;" shows otherwise and that is true as a list of long running querys showed. Is there a way to make it behave like the interactive version ? Now it is a bit confusing for everyone. re, wh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: Re: signal handling in mysql cli
Hello Sebastein, On 2014-12-03 17:25, Sebastien FLAESCH wrote: Hi all, I have a similar question regarding KILL QUERY usage: We have a C client program using libmysqlclient.so, it is a single-threaded program. When running a long query, how can I send the KILL QUERY command when a SIGINT (CTRL-C) is caught? (of course we implement a signal handler, so we keep the control) = Is is safe to establish a new connection to the server in the signal handler, using mysql_init() + mysql_real_connect(), and execute the KILL QUERY with the mysql thread id I got from the initial mysql_init() / mysql_real_connect()? I made some tests, and it seems to work fine, the long query returns SQL error -1317: Query execution was interrupted (which is expected) We want to support SQL interruption properly, so please someone from the dev team, give me a clear answer... I don't want to use a side effect or undocumented feature. Issuing KILL QUERY from a separate connection is the way to do it. We do not currently have any CAPI abort function - introducing such functionality would require changes on the protocol level. Doing all this stuff in a signal handler is certainly risky... no? Whether it is safe to do I/O extensive work inside a signal handler is another topic, not related to MySQL client library. Depends on your particular case and how careful you are about writing handler code which can be executed at any time. One safe option is that signal handler only sets a flag informing about the signal and then either the main logic of your application detects the flag and performs KILL operation, or you have a separate dedicated thread to do this job. Rafal (from Connectors/C team @ MySQL) I could not find the information in the documentation (mysql_real_connect). I wish there would be an API like mysql_cancel_query(), similar to Oracle's OCI OCIBreak(). Thanks! Seb On 12/02/2014 05:13 PM, walter harms wrote: hi list, when i use CTRL-C to break a query that works fine in interactive mode. mysql select sleep(10) ; ^CCtrl-C -- sending KILL QUERY 24289 to server ... Ctrl-C -- query aborted. +---+ | sleep(10) | +---+ +---+ 1 row in set (0.86 sec) but when i use the noninteractive mode timeout 5 mysql -BAN -e select now(); select sleep (100) ; select now() i looks like that but show full processlist; shows otherwise and that is true as a list of long running querys showed. Is there a way to make it behave like the interactive version ? Now it is a bit confusing for everyone. re, wh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
Thank you Rafal for your answer!
We do not use threads, it's a single threaded program. So we cannot set
a flag to exec the KILL QUERY later on in another thread.
The only flag I use is to avoid subsequent KILL QUERY attempts when one
has already been sent...
The code in the sigint handler is quite simple, basic mysql_ calls and
debug messages (fprintfs):
if (currConn currConn-canInterrupt) {
char cmd[50];
MYSQL *h;
currConn-canInterrupt = 0;
h = mysql_init(NULL);
if (!mysql_real_connect(h, currConn-c_host, currConn-c_user,
currConn-c_auth, currConn-c_dbnm,
currConn-c_port, currConn-c_sock,
CLIENT_FOUND_ROWS)) {
DBGMSGINT(2, Interrupt: Could not connect to database: %d.,
mysql_errno(h));
return 0;
}
sprintf(cmd, KILL QUERY %ld, currConn-threadId);
if (mysql_query(h, cmd) != 0) {
DBGMSGSTR(2, Could not execute %s., cmd);
}
mysql_close(h);
}
Should do the job...
Thanks.
Seb
On 12/12/2014 12:12 PM, rafal somla wrote:
Hello Sebastein,
On 2014-12-03 17:25, Sebastien FLAESCH wrote:
Hi all,
I have a similar question regarding KILL QUERY usage:
We have a C client program using libmysqlclient.so, it is a
single-threaded program.
When running a long query, how can I send the KILL QUERY command when a
SIGINT (CTRL-C)
is caught? (of course we implement a signal handler, so we keep the
control)
= Is is safe to establish a new connection to the server in the signal
handler, using
mysql_init() + mysql_real_connect(), and execute the KILL QUERY with the
mysql thread
id I got from the initial mysql_init() / mysql_real_connect()?
I made some tests, and it seems to work fine, the long query returns SQL
error -1317:
Query execution was interrupted (which is expected)
We want to support SQL interruption properly, so please someone from the
dev team,
give me a clear answer... I don't want to use a side effect or
undocumented feature.
Issuing KILL QUERY from a separate connection is the way to do it. We do not currently
have any CAPI abort function - introducing such functionality
would require changes on the protocol level.
Doing all this stuff in a signal handler is certainly risky... no?
Whether it is safe to do I/O extensive work inside a signal handler is another
topic, not related to MySQL client library. Depends on your particular
case and how careful you are about writing handler code which can be executed
at any time. One safe option is that signal handler only sets a flag
informing about the signal and then either the main logic of your application
detects the flag and performs KILL operation, or you have a separate
dedicated thread to do this job.
Rafal (from Connectors/C team @ MySQL)
I could not find the information in the documentation (mysql_real_connect).
I wish there would be an API like mysql_cancel_query(), similar to
Oracle's OCI
OCIBreak().
Thanks!
Seb
On 12/02/2014 05:13 PM, walter harms wrote:
hi list,
when i use CTRL-C to break a query that works fine in interactive mode.
mysql select sleep(10) ;
^CCtrl-C -- sending KILL QUERY 24289 to server ...
Ctrl-C -- query aborted.
+---+
| sleep(10) |
+---+
+---+
1 row in set (0.86 sec)
but when i use the noninteractive mode
timeout 5 mysql -BAN -e select now(); select sleep (100) ; select
now()
i looks like that but show full processlist; shows otherwise and
that is true
as a list of long running querys showed.
Is there a way to make it behave like the interactive version ?
Now it is a bit confusing for everyone.
re,
wh
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
Nobody concerned by this case? Simple question: is it safe or not to do a KILL QUERY in a SIGINT signal handler? S. On 12/03/2014 05:25 PM, Sebastien FLAESCH wrote: Hi all, I have a similar question regarding KILL QUERY usage: We have a C client program using libmysqlclient.so, it is a single-threaded program. When running a long query, how can I send the KILL QUERY command when a SIGINT (CTRL-C) is caught? (of course we implement a signal handler, so we keep the control) = Is is safe to establish a new connection to the server in the signal handler, using mysql_init() + mysql_real_connect(), and execute the KILL QUERY with the mysql thread id I got from the initial mysql_init() / mysql_real_connect()? I made some tests, and it seems to work fine, the long query returns SQL error -1317: Query execution was interrupted (which is expected) We want to support SQL interruption properly, so please someone from the dev team, give me a clear answer... I don't want to use a side effect or undocumented feature. Doing all this stuff in a signal handler is certainly risky... no? I could not find the information in the documentation (mysql_real_connect). I wish there would be an API like mysql_cancel_query(), similar to Oracle's OCI OCIBreak(). Thanks! Seb On 12/02/2014 05:13 PM, walter harms wrote: hi list, when i use CTRL-C to break a query that works fine in interactive mode. mysql select sleep(10) ; ^CCtrl-C -- sending KILL QUERY 24289 to server ... Ctrl-C -- query aborted. +---+ | sleep(10) | +---+ +---+ 1 row in set (0.86 sec) but when i use the noninteractive mode timeout 5 mysql -BAN -e select now(); select sleep (100) ; select now() i looks like that but show full processlist; shows otherwise and that is true as a list of long running querys showed. Is there a way to make it behave like the interactive version ? Now it is a bit confusing for everyone. re, wh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
- Original Message - From: Sebastien FLAESCH s...@4js.com Subject: Re: signal handling in mysql cli Nobody concerned by this case? Simple question: is it safe or not to do a KILL QUERY in a SIGINT signal handler? I don't see what the extra risk would be as opposed to doing so in any other thread of your code, or in another program entirely. As long as you're sure about which particular query you're shooting in the face, it should be perfectly fine. -- Unhappiness is discouraged and will be corrected with kitten pictures. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
Am 02.12.2014 18:31, schrieb Johan De Meersman: - Original Message - From: wharms wha...@bfs.de Subject: signal handling in mysql cli when i use CTRL-C to break a query that works fine in interactive mode. but when i use the noninteractive mode i looks like that but show full processlist; shows otherwise and that is true This may sound silly, but you're basically saying that you can't interact with it while in non-interactive mode... My understanding may be wrong, but this is how it works in my mind: * when in the client, the client intercepts the ctrl-c and interprets it as kill this query on the server. * when in non-interactive mode, the client is not actually reading your keypresses. Thus, the ctrl-c gets intercepted by the *shell*, which does exactly what you ask by killing the MySQL client process. Now, if the mysql client does not explicitly intercept kill signals - which, honestly, most programs have no reason to - that means it doesn't even get a chance to send the cancel that query order, it just dies. A lot of debate could be put in on wether or not the client needs to intercept and send a cancel, but personally I'm leaning towards no, it doesn't - if you don't have a transaction open, or don't even have a transactional engine (which used to be default!) cancelling would leave your database in an inconsistent state. Much better, then, to allow whatever query you sent to continue, on the off chance that it's something that it does not damage, or only does things that you can reverse afterwards. If you really want to kill that query, you always have the option of logging into the server and shooting it yourself. In Unix pressing CTRL-C should be the same a sending a INTR. You can see this with stty -a what key is map to what signal. NTL the question is: how can the user stop the query what was started with a script ? re, wh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
Hi all, I have a similar question regarding KILL QUERY usage: We have a C client program using libmysqlclient.so, it is a single-threaded program. When running a long query, how can I send the KILL QUERY command when a SIGINT (CTRL-C) is caught? (of course we implement a signal handler, so we keep the control) = Is is safe to establish a new connection to the server in the signal handler, using mysql_init() + mysql_real_connect(), and execute the KILL QUERY with the mysql thread id I got from the initial mysql_init() / mysql_real_connect()? I made some tests, and it seems to work fine, the long query returns SQL error -1317: Query execution was interrupted (which is expected) We want to support SQL interruption properly, so please someone from the dev team, give me a clear answer... I don't want to use a side effect or undocumented feature. Doing all this stuff in a signal handler is certainly risky... no? I could not find the information in the documentation (mysql_real_connect). I wish there would be an API like mysql_cancel_query(), similar to Oracle's OCI OCIBreak(). Thanks! Seb On 12/02/2014 05:13 PM, walter harms wrote: hi list, when i use CTRL-C to break a query that works fine in interactive mode. mysql select sleep(10) ; ^CCtrl-C -- sending KILL QUERY 24289 to server ... Ctrl-C -- query aborted. +---+ | sleep(10) | +---+ +---+ 1 row in set (0.86 sec) but when i use the noninteractive mode timeout 5 mysql -BAN -e select now(); select sleep (100) ; select now() i looks like that but show full processlist; shows otherwise and that is true as a list of long running querys showed. Is there a way to make it behave like the interactive version ? Now it is a bit confusing for everyone. re, wh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
signal handling in mysql cli
hi list, when i use CTRL-C to break a query that works fine in interactive mode. mysql select sleep(10) ; ^CCtrl-C -- sending KILL QUERY 24289 to server ... Ctrl-C -- query aborted. +---+ | sleep(10) | +---+ +---+ 1 row in set (0.86 sec) but when i use the noninteractive mode timeout 5 mysql -BAN -e select now(); select sleep (100) ; select now() i looks like that but show full processlist; shows otherwise and that is true as a list of long running querys showed. Is there a way to make it behave like the interactive version ? Now it is a bit confusing for everyone. re, wh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
Re: signal handling in mysql cli
- Original Message - From: wharms wha...@bfs.de Subject: signal handling in mysql cli when i use CTRL-C to break a query that works fine in interactive mode. but when i use the noninteractive mode i looks like that but show full processlist; shows otherwise and that is true This may sound silly, but you're basically saying that you can't interact with it while in non-interactive mode... My understanding may be wrong, but this is how it works in my mind: * when in the client, the client intercepts the ctrl-c and interprets it as kill this query on the server. * when in non-interactive mode, the client is not actually reading your keypresses. Thus, the ctrl-c gets intercepted by the *shell*, which does exactly what you ask by killing the MySQL client process. Now, if the mysql client does not explicitly intercept kill signals - which, honestly, most programs have no reason to - that means it doesn't even get a chance to send the cancel that query order, it just dies. A lot of debate could be put in on wether or not the client needs to intercept and send a cancel, but personally I'm leaning towards no, it doesn't - if you don't have a transaction open, or don't even have a transactional engine (which used to be default!) cancelling would leave your database in an inconsistent state. Much better, then, to allow whatever query you sent to continue, on the off chance that it's something that it does not damage, or only does things that you can reverse afterwards. If you really want to kill that query, you always have the option of logging into the server and shooting it yourself. /johan -- Unhappiness is discouraged and will be corrected with kitten pictures. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/mysql
