> that seems wrong, for those which are symlinks (ie. 0->6) can you also > readlink and print that out too please? > > it seems like some fd's are beinh held open > > as a work around you can loop over these fd's (say 2 though 255) and > set FD_CLOEXEC (lots of things do this, it's a but of a hack and not > entirely reliable)
I repeated the test and ran readlink for each fd as you suggested. Here are the results before SQLite: . .. 0 /dev/console1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 1 /dev/console1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 2 /dev/console1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 3 /proc/130/fd1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 After SQLite: . /proc/130/fd1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 .. /proc/130/fd1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 0 /dev/console1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 1 /dev/console1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 2 /dev/console1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 3 /proc/130/fd1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 5 pipe:[120]fd1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 6 pipe:[120]fd1 ¶g¤T 1¤ë 2 04:54:50 CST 2008 I am still relatively new to Linux so some of the commands you mentioned are unfamiliar to me. I hope this output is what you wanted. Could you explain how I can set FD_CLOEXEC? If its unreliable then I really can't use it permanently but it may be a useful test to run. Thanks Chris _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
