makemap: error opening type hash map access: Invalid argument
in [Sendmail]
|
Prev: SYSERR(root): buildaddr: unknown mailer relay
Next: makemap: error opening type hash map access: Invalidargument
From: JP on 15 Nov 2009 19:45 i'm installing a new machine, and i ran into something that i haven't seen before, any help and pointers appreciated; let me know if more info is needed OS: linux sendmail-8.14.3 and db-4.8.24 (both compiled from source) # makemap hash access < access access.db: Invalid type 5 specified makemap: error opening type hash map access: Invalid argument # makemap -l hash btree # ldd /usr/local/sbin/makemap linux-gate.so.1 => (0xb77b8000) libdb-4.8.so => /usr/local/lib/libdb-4.8.so (0xb7644000) libresolv.so.2 => /lib/libresolv.so.2 (0xb762e000) libcrypt.so.1 => /lib/libcrypt.so.1 (0xb75fc000) libnsl.so.1 => /lib/libnsl.so.1 (0xb75e3000) libdl.so.2 => /lib/libdl.so.2 (0xb75df000) libc.so.6 => /lib/libc.so.6 (0xb747f000) libpthread.so.0 => /lib/libpthread.so.0 (0xb7466000) /lib/ld-linux.so.2 (0xb77b9000) J.
From: mikea on 16 Nov 2009 10:30 JP <jurgenph(a)gmail.com.com> wrote in <qI2dnZST66U1AZ3WnZ2dnUVZ_gydnZ2d(a)speakeasy.net>: > i'm installing a new machine, and i ran into something that i haven't > seen before, any help and pointers appreciated; > > let me know if more info is needed > > OS: linux > sendmail-8.14.3 and db-4.8.24 (both compiled from source) > > # makemap hash access < access > access.db: Invalid type 5 specified > makemap: error opening type hash map access: Invalid argument 1) What does `file access access.db` produce? 2) Can you isolate the failure to a specific line in access? > # makemap -l > hash > btree > > # ldd /usr/local/sbin/makemap > linux-gate.so.1 => (0xb77b8000) > libdb-4.8.so => /usr/local/lib/libdb-4.8.so (0xb7644000) > libresolv.so.2 => /lib/libresolv.so.2 (0xb762e000) > libcrypt.so.1 => /lib/libcrypt.so.1 (0xb75fc000) > libnsl.so.1 => /lib/libnsl.so.1 (0xb75e3000) > libdl.so.2 => /lib/libdl.so.2 (0xb75df000) > libc.so.6 => /lib/libc.so.6 (0xb747f000) > libpthread.so.0 => /lib/libpthread.so.0 (0xb7466000) > /lib/ld-linux.so.2 (0xb77b9000) -- Mike Andrews, W5EGO mikea(a)mikea.ath.cx Tired old sysadmin
From: JP on 16 Nov 2009 13:03 mikea wrote: >> # makemap hash access < access >> access.db: Invalid type 5 specified >> makemap: error opening type hash map access: Invalid argument > > 1) What does `file access access.db` produce? > > 2) Can you isolate the failure to a specific line in access? i'm starting from scratch, there is currently only one line in the access file, once everything is up and running i'll populate it with more entries. -rw-r--r-- 1 root root 18 2009-11-16 09:58 access -rw-r----- 1 root root 0 2009-11-16 09:58 access.db # file access* access: ASCII text access.db: empty # cat access 64.81.52.5 RELAY in all the years, i've never been this puzzled by a problem... it's been two days and countless recompiles later, and still not a clue why this is happening. i wonder if i should try an older version of berkeleydb instead... J.
From: Loki Harfagr on 16 Nov 2009 13:15 Mon, 16 Nov 2009 10:03:57 -0800, JP did cat : > mikea wrote: > >>> # makemap hash access < access >>> access.db: Invalid type 5 specified >>> makemap: error opening type hash map access: Invalid argument >> >> 1) What does `file access access.db` produce? >> >> 2) Can you isolate the failure to a specific line in access? > > i'm starting from scratch, there is currently only one line in the > access file, once everything is up and running i'll populate it with > more entries. > > -rw-r--r-- 1 root root 18 2009-11-16 09:58 access -rw-r----- 1 > root root 0 2009-11-16 09:58 access.db > > # file access* > access: ASCII text > access.db: empty > > # cat access > 64.81.52.5 RELAY > > > in all the years, i've never been this puzzled by a problem... it's been > two days and countless recompiles later, and still not a clue why this > is happening. > > i wonder if i should try an older version of berkeleydb instead... > > > J. I know you're already thinking about it but just for info what gives the DB inborn check?: # sendmail -bv -d0.1 | grep DB
From: JP on 16 Nov 2009 14:14
Loki Harfagr wrote: > I know you're already thinking about it but just for info > what gives the DB inborn check?: > # sendmail -bv -d0.1 | grep DB # sendmail -bv -d0.1 | grep DB NAMED_BIND NETINET NETUNIX NEWDB PIPELINING SCANF USERDB XDEBUG J. |