search for: unique_len

...ret = fcntl(lockfile_fd, F_SETLK, &lck); if (ret == -1) { ret = errno; DEBUG(1, ("%s: fcntl failed: %s\n", __func__, strerror(ret))); goto fail_close; } DEBUG(1, ("unique is %u\n", unique)); unique_len = snprintf(buf, sizeof(buf), "%ju\n", (uintmax_t)unique); DEBUG(1, ("unique_len is: %i\n", unique_len)); /* shorten a potentially preexisting file */ ret = ftruncate(lockfile_fd, unique_len); if (ret == -1) { ret = errno;...

I am trying to see if Samba 4.2.0rc4 will run on a Solaris 10 i386 system. After getting it to build, startup fails. Starting it up as smbd -i -d 10 The output indicates that everything is starting normally until the following is output: messaging_dgm_lockfile_create: ftruncate failed: Invalid argument messaging_dgm_init: messaging_dgm_create_lockfile failed: Invalid argument messaging_dgm_init

...t; not do it.... > > Volker You guys sure work at odd hours. I should have mentioned that when I put in the DEBUG lines I had to not include the %j modifier as the DEBUG produces no output when it is there. I just tried removing the j from the snprintf line leaving %u and now the value of unique_len is comming out apparently correctly. The largest value of unique that I have seen is 4125756149 and the largest value of unique_len is 11. The Solaris man page for snprintf claims that the %j modifier should work. Tom Schulz Applied Dynamics Intl. schulz at adi.com

...have access to Solaris >> right now) patch *might* already do it for you. Can you give >> it a try? >> >> Thanks, >> >> Volker > > That seems to work. I put the j back into the sprintf line. I now get > values of unique such as 5477310573428528709 and unique_len of 20. I am somewhat surprised that your replacement snprintf is being used as Solaris does have that function. Just for your information, here is what configure has to say about *printf. Checking for printf : ok Checking for dprintf : not f...