Hi, On 9/15/2006 2:50 PM, Fernando Lima wrote: > Hello, > > I'm performing tests of backup/restore using Bacula. > > The software is fantastic and was working better than my current > solution "Arcserve". > > Two days ago, I tried a test of backup/restore of one special server. > This server store half a million pdf files in one directory !
Bad in itself... > The file > name structure is the following (Without "): > > "XXX1111111 XXX9999XXX 00 00000002 ".pdf > > All files have 52 characters. Usually the only difference between them > are the positions 4-10 and 51 (The numbers 1 and 2 above). > > The backup worked without problems. But when I tried the restore > operation my problem began. > > From the restore command until the tree prompt it took 13 hours ! Yup. > I did some research on the list mail archive and found some discussion > about the same problem one year ago . I found two solutions. The first > solution was to check the mysql index. I checked all of them and they > looked fine. the second solution was a code change, Reinier Haasjes > suggested the problem is in the binary search ( > http://www.mail-archive.com/bacula-users@lists.sourceforge.net/msg03358.html > ) and recommended another index library. I can't do that ! > > I would like some ideas to implement a workaround for this problem. As far as I can tell it should be possible to create or modify the indexes for that kind of fileset. There were discussions detailing this, one rather recently where it was suggested to modify the index length to cover the whole entry. This was about the path table, but for the Filename table, it should work similar. Although I see that, at least here, the index covers 255 characters of file name data which should be nough for your case. Probably, if the path your files are in, is longer than the default 50 characters for the index on pathname it might help setting this index to cover the whole path length. > The only solution I found until now, was backup that server using a > special Pool with the parameter Catalog Files = no. That will make restoring individual files rather difficult. > I'm trying a restore using bextract with an exclude list but after two > hours nothing happend yet. (Message "Reading..."). Well, probably bextract has to read some time before it reaches any file that should be processed. > > My environment is: > > Pentium III - 550 Mhz > 1 Gb Memory > 3 disk concat total 450gb > 1 Tape SDLT > Bacula 1.38.11 > Director + Storage in the same server > Linux Fedora Core 4 > Mysql 5.0.24 Hope this helps, Arno > thanks in advance, > > Silva. > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job easier > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > > > ------------------------------------------------------------------------ > > _______________________________________________ > Bacula-users mailing list > Bacula-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bacula-users -- IT-Service Lehmann [EMAIL PROTECTED] Arno Lehmann http://www.its-lehmann.de ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
