flex -o core/lex.yy.c core/cfg.lex CC (gcc) [kamailio] core/lex.yy.o CC (gcc) [kamailio] core/cfg.tab.o LD (gcc) [kamailio] kamailio CC (gcc) [M db_mysql.so] km_val.o CC (gcc) [M db_mysql.so] my_fld.o In file included from my_fld.c:22:0: my_fld.h:37:2: error: unknown type name ‘my_bool’ my_bool is_null; ^~~~~~~ ../../Makefile.rules:100: recipe for target 'my_fld.o' failed make[2]: *** [my_fld.o] Error 1 Makefile:511: recipe for target 'modules' failed make[1]: *** [modules] Error 1 make[1]: Leaving directory '/usr/local/src/kamailio-5.1/kamailio/src' Makefile:27: recipe for target 'all' failed make: *** [all] Error 2
We are building master branch every night for [bionic](https://kamailio.sipwise.com/job/kamailiodev-nightly-binaries/architecture=a...) with no problems.
``` /usr/bin/make -C src/ every-module make[2]: Entering directory '/tmp/buildd/kamailio-5.2.0~dev6+0~20180726010658.1165+bionic/src' CC (gcc) [M db_mysql.so] km_dbase.o CC (gcc) [M db_mysql.so] km_my_con.o CC (gcc) [M db_mysql.so] km_val.o CC (gcc) [M db_mysql.so] my_fld.o CC (gcc) [M db_mysql.so] km_row.o CC (gcc) [M db_mysql.so] my_con.o CC (gcc) [M db_mysql.so] km_res.o CC (gcc) [M db_mysql.so] my_cmd.o CC (gcc) [M db_mysql.so] db_mysql.o CC (gcc) [M db_mysql.so] my_res.o CC (gcc) [M db_mysql.so] my_uri.o CC (gcc) [M db_mysql.so] km_db_mysql.o make[4]: 'libsrdb2.so.1.0' is up to date. make[4]: 'libsrdb1.so.1.0' is up to date. LD (gcc) [M db_mysql.so] db_mysql.so make[2]: Leaving directory '/tmp/buildd/kamailio-5.2.0~dev6+0~20180726010658.1165+bionic/src' ```
And there are no related changes on db_mysql AFAICT I would say you are missing dev libraries
I guess it was used with mysql 8.0.1, where my_bool type was removed -- patch above should be the fix, but it would require C99 compatible compiler. If master gets compiled/packaged without issues on the systems we build, then it should be ok.
@limitall could you test commit 1736723cc2a15bb75c26711675712a966161b722 or current master if now master builds correctly, then please close ticket.
I am closing it, if something new appears and still a problem, reopen.
Closed #1602.
Just for reference: the fix referenced above caused a similar issue for MariaDB, it is now fixed in commit 9f459dccf84b in master.
-
Daniel-Constantin Mierla -
Henning Westerholt
-
limitall
-
sergey-safarov
-
Victor Seva