On 2007-02-15 15:59, Chris Hoogendyk wrote: > NOTE: I was just writing a wrapper script for ufsdump on Solaris 9 to > incorporate fssnap snapshots and a couple of other things (under > 2.5.1p3). I noticed that my script was being called with the options > "0SSsf". ufsdump did not seem to mind, and worked without any hangups. I > also tried manually doing it with "0Ssf" and with "0SSSsf" and it made > no difference. > > However, if this is an indication that amanda wasn't accounting for > Solaris and ufsdump, and that it would substitute "0Shsf" when patched > to work "correctly", then it will break on Solaris, which has no "h" > option. I hope somone can look at the code and see that it is handling > the various OS differences correctly. The double "SS" and your note > below makes it sound like amanda might not be accounting for Solaris > properly.
The patch below changes line 1193 into "PARAM_HONOR_NODUMP". If you did the trouble to look there, then you see a few lines bove (lines 1186-1190): 1186 # ifdef HAVE_HONOR_NODUMP 1187 # define PARAM_HONOR_NODUMP "h" 1188 # else 1189 # define PARAM_HONOR_NODUMP "" 1190 # endif 1191 dumpkeys = vstralloc(level_str, 1192 PARAM_DUMP_ESTIMATE, 1193 PARAM_HONOR_NODUMP, 1194 "s", "f", NULL); So, yes, if you do run ./Configure on a Solaris machine, then the flag will be correctly omitted. > > Thanks, NP. [...] > > Jean-Louis Martineau wrote: >> Chris, >> >> The attached patch fix the problem. >> The problem is the two 'S', one of them should be 'h'. >> >> The patch is already in the CVS. >> >> Jean-Louis >> >> Chris Stenton wrote: >>> Not sure if this is to do with the FreeBSD port but upgrading to >>> 2.5.1p3 from p2 and the dump estimate has stopped working. Here is >>> the debug output from client sendsize. >>> >>> sendsize[43898]: time 1.025: calculating for device /dev/da0s1a with ufs >>> sendsize[43898]: time 1.025: running "/sbin/dump 1SSsf 0 1048576 - >>> /dev/da0s1a" >>> sendsize[43898]: time 1.026: running /usr/local/libexec/amanda/killpgrp >>> sendsize[43898]: time 1.030: dump: tape size must be between 1 and 0 >>> sendsize[43898]: time 1.030: ..... >>> >>> >>> not sure why the extra argument "0" has been added. Any ideas? >>> >>> Thanks >>> >>> >>> Chris >>> >> >> ------------------------------------------------------------------------ >> >> diff -u -r --show-c-function --new-file >> --exclude-from=/home/martinea/src.orig/amanda.diff >> --ignore-matching-lines='$Id:' >> amanda-2.5.1p3.new/client-src/sendsize.c >> amanda-2.5.1p3.new.krb4/client-src/sendsize.c >> --- amanda-2.5.1p3.new/client-src/sendsize.c 2006-12-22 >> 09:42:42.000000000 -0500 >> +++ amanda-2.5.1p3.new.krb4/client-src/sendsize.c 2007-02-08 >> 15:13:57.000000000 -0500 >> @@ -1190,7 +1190,7 @@ getsize_dump( >> # endif >> dumpkeys = vstralloc(level_str, >> PARAM_DUMP_ESTIMATE, >> - PARAM_DUMP_ESTIMATE, >> + PARAM_HONOR_NODUMP, >> "s", "f", NULL); >> >> # ifdef HAVE_DUMP_ESTIMATE >> -- Paul Bijnens, xplanation Technology Services Tel +32 16 397.511 Technologielaan 21 bus 2, B-3001 Leuven, BELGIUM Fax +32 16 397.512 http://www.xplanation.com/ email: [EMAIL PROTECTED] *********************************************************************** * I think I've got the hang of it now: exit, ^D, ^C, ^\, ^Z, ^Q, ^^, * * F6, quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, * * stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt, abort, hangup, * * PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e, kill -1 $$, shutdown, * * init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... * * ... "Are you sure?" ... YES ... Phew ... I'm out * ***********************************************************************
