Re: [Bacula-users] Anyone written any handy queries (query.sql)???
John Lockard wrote, sometime around 11/08/09 19:39: I have modified my query.sql to include some queries that I use frequently and I thought maybe someone else would find them useful additions. Also, I was wondering if anyone had queries which they find useful and would like to share. Some of mine may only work on PostgreSQL - I dunno.. :List the most recent full backups SELECT DISTINCT jobname, endtime, volumename from ( SELECT lastfullbackups.JobId, JobName, ScheduledTime, StartTime, EndTime, volumename FROM ( SELECT jobid AS JobId, job.name AS JobName, job.schedtime AS ScheduledTime, job.starttime AS StartTime, job.endtime AS EndTime FROM ( SELECT name as Job, max(endtime) AS EndTime FROM job INNER JOIN jobinfo ON job.name = JI_jobname WHERE type = 'B' AND level = 'F' AND jobstatus = 'T' AND ji_old = false GROUP BY name ) lastfullbackups LEFT OUTER JOIN job ON lastfullbackups.Job = job.name AND lastfullbackups.EndTime = job.endtime ) lastfullbackups LEFT OUTER JOIN jobmedia ON lastfullbackups.jobid = jobmedia.jobid LEFT OUTER JOIN media ON jobmedia.mediaid = media.mediaid ) foo :List last 20 Full Backups for a named job *Enter Job name: SELECT DISTINCT Job.JobId,Client.Name AS Client,Job.StartTime,JobFiles,JobBytes, VolumeName FROM Client INNER JOIN Job on Client.ClientId = Job.ClientId INNER JOIN JobMedia ON Job.JobId = JobMedia.JobId INNER JOIN Media on JobMedia.MediaId=Media.MediaId WHERE Job.Name='%1' AND Level='F' AND JobStatus='T' ORDER BY Job.StartTime DESC LIMIT 20; Also, I wanted to know which tapes needed to be offsite when I was running a full differential combination. To do this, I created a table listing all the jobs and whether they should be included or not as I have quite a few old jobs in the catalog which I don't really care about: CREATE TABLE jobinfo ( ji_primary SERIAL, ji_old boolean DEFAULT FALSE, ji_jobname varchar(45) ); List the last completed full backup for each job, and the tape it is on: CREATE VIEW lastfullbackuptapes AS SELECT DISTINCT foo2.jobname, foo2.endtime, foo2.volumename FROM ( SELECT lastfullbackups.jobid, jobname, scheduledtime, starttime, endtime, volumename FROM ( SELECT jobid, lastfullbackups.job AS jobname, job.schedtime AS scheduledtime, job.starttime, job.endtime FROM ( SELECT ji_jobname AS job, max(endtime) AS endtime FROM jobinfo LEFT JOIN ( SELECT job.name, job.endtime FROM job WHERE job.type = 'B'::bpchar AND job.level = 'F' AND job.jobstatus = 'T') fulljobs ON jobinfo.ji_jobname = fulljobs.name WHERE ji_old = false GROUP BY ji_jobname) lastfullbackups LEFT JOIN job ON lastfullbackups.job = job.name AND lastfullbackups.endtime = job.endtime) lastfullbackups LEFT JOIN jobmedia ON lastfullbackups.jobid = jobmedia.jobid LEFT JOIN media ON jobmedia.mediaid = media.mediaid) foo2 ORDER BY foo2.jobname, foo2.endtime, foo2.volumename; List all completed differential backups since the last full backup for each job and the tape they are on (uses the above view): CREATE VIEW lastdiffbackuptapes AS SELECT DISTINCT res.name, res.endtime, media.volumename FROM lastfullbackuptapes JOIN ( SELECT job.jobid, job.job, job.name, job.type, job.level, job.clientid, job.jobstatus, job.schedtime, job.starttime, job.endtime, job.jobtdate, job.volsessionid, job.volsessiontime, job.jobfiles, job.jobbytes, job.joberrors, job.jobmissingfiles, job.poolid, job.filesetid, job.purgedfiles, job.hasbase FROM job WHERE job.type = 'B' AND job.jobstatus = 'T' AND job.level = 'D') res ON lastfullbackuptapes.jobname = res.name AND lastfullbackuptapes.endtime res.starttime LEFT JOIN jobmedia ON res.jobid = jobmedia.jobid LEFT JOIN media ON jobmedia.mediaid = media.mediaid ORDER BY res.name, res.endtime, media.volumename; List all tapes which should be offsite in order to maintain the most up to date complete backup (for each job the most recent full and all subsequent differentials): SELECT DISTINCT res.volumename FROM ( SELECT lastfullbackuptapes.volumename FROM lastfullbackuptapes UNION SELECT lastdiffbackuptapes.volumename FROM lastdiffbackuptapes) res ORDER BY res.volumename; -- Russell Howe, IT Manager. rh...@bmtmarinerisk.com BMT Marine Offshore Surveys Ltd
Re: [Bacula-users] How to automate Windows installer for a large number of hosts?
Brian Debelius wrote, sometime around 26/01/09 17:05: You could try WinInstall to do this. http://www.scalable.com/WinInstall.aspx Also take a look at AutoIt - I think the Unattended project (http://unattended.sf.net/) uses it quite a bit. That said, NSIS isn't very complicated - I remember using it ages ago to package openphone for internal use -- Russell Howe, IT Manager. rh...@bmtmarinerisk.com BMT Marine Offshore Surveys Ltd. 新年快乐! -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Clustering the win32 FD
Robert LeBlanc wrote: I was just thinking about this. And this might be what you described below. Create a config with with same password for blue and green. Create three jobs and two filesets. Have one job back-up blue (except for data disks), one job back-up green (except for data disks) and then another job that points to the cluster IP address that only backs up the data. One fileset can be used for blue and green and the other for the data. So you would have three clients for two machines. The only hang-up I can see is that the fd name on the dir and client would not match, but I'm not sure if it does password plus name or just password for auth. It would required 0.0.0.0:9102 or you could possibly get away with multiple IP address statements in the fd or comma separated addresses. That's the way I'm currently doing it (with 0.0.0.0:9102) listening addresses (the default). But, I'd really like it as a clustered service, and it shouldn't be that hard to do. I'll have another stab at it next week. -- Russell Howe [EMAIL PROTECTED] - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Clustering the win32 FD
Robert LeBlanc wrote:
I don't think you want to cluster the bacula FD. Each node of the cluster is
very unique and if you try to fail over the back-up program, you will get
some really strange results. All of a sudden a computer that you think has
one name and identification and config in the cluster now has another. This
is assuming that you are backing up the entire machine. If you are only
backing up the data disk, this might be ok. In such a situation you will
probably want a client on each node and then one that fails over.
Yes, this is the configuration I'm looking for.
Let's take a clustered file server, for a simple example. Maybe with two
cluster groups.
Server #1 called 'green'
Server #2 called 'blue'
Cluster group called 'User Dept filestores' with a disk or two,
network name ('usrdeptsrv'), IP address (192.168.0.100) and file shares
('homedrives', 'deptstore').
Cluster group called 'Archive storage' with another disk or two, and
again a network name ('archivesrv'), IP address (192.168.0.200) and file
share ('archive')
Plus a group for the quorum, which isn't really relevant to the backup
scenario.
Now, what I'm aiming for is to be able to back up green and blue's
system state and configuration, so I can install a filedaemon on them,
calling them something like green-fd and blue-fd.
However, I can't really use those to back up the clustered file stores,
as I don't know whether green or blue will have the drives. I could just
run the backups for the clustered drives on both the servers, and the
one which has the drives will succeed and the one without the drives
will fail, but it's not an ideal scenario.
The other thing you can do is to set green-fd and blue-fd with the same
password, then create another Client resource in the director's config
called (say) 'archivesrv-fd' with the IP address which is linked to the
archive storage. That way you can be sure that the connection is being
made to the node which has the files, and you'll be able to authenticate
as the credentials are the same for both nodes.
Still, that's not ideal and it isn't really obvious what's going on.
What
I've done for clusters is just put the client on each node and which ever
one has the data disk backs it up. All of our clusters have fail back so the
data usually resides on the primary node, but if it doesn't then I can
restore it from the standby node job if the job ran when the standby node
owned the disk. You can easily restore to the standby node if the primary
node is dead. Honestly, I think you are go through a lot more work than it
is worth. If you have a really good reason, I'd like to know.
Well, my reason is that it really shouldn't be hard, and it should be
the Right Way(tm) to do it.
Install green-fd to listen on green's IP address, blue-fd to listen on
blue's IP address (instead of having them listen on 0.0.0.0:9102!), and
then one fd per cluster group, failing over with the cluster. It keeps a
clean separation between the director and the config of the servers.
It looks like it should be relatively easy, but I think maybe with my
messing around trying to get it to work, I may have left trails of
destruction in the parts of the registry to do with service
registrations and got Windows all confuddled.
--
Russell Howe
[EMAIL PROTECTED]
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Clustering the win32 FD
Has anybody ever run bacula-fd under the Windows cluster service? I've been trying to create it as a Generic Service without a whole lot of success. Still, I haven't exhausted all troubleshooting options as yet, so I might get it going... -- Russell Howe [EMAIL PROTECTED] - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Backing up MS SQL server with Bacula?
Robert LeBlanc wrote, sometime around 11/10/07 12:03: We put MS SQL in full recovery mode, then have a maintenance schedule in MS SQL that does a full back-up. We exclude the actual data files and include the log files (full and transaction). We have the schedule run about 15 min before bacula runs on the client. I think the RunBefore Client would be a better option though. I currently have a maintenance plan which does a consistency check and then database backup followed by log backup (the latter is needed if you don't want your log files to grow to infinite size when a database is in 'Full' recovery mode). This currently runs asynchronously w.r.t. bacula, so I just make sure that I schedule the bacula job to run at a time by which the database server will have dumped its backups. I also have a client run before job script (shell script executed with cygwin), which prunes backup dump files older than a week, and gzips the remainder (SQL Server backup files gzip rather well, I find). I think I trust gzip to do a better job than the tape drive's hardware compression. If you look at the maintenance plan, all it does is call a stored procedure with some arguments. It should be trivial to call this from the client run before job too, making the whole process synchronous. I also saw someone having some success running backups over named pipes and streaming them directly to tape, avoiding the middle step of dumping databases to files before backing up. I like having an online copy of the database backups though, as it makes restoring in the common case (something went wrong and we need last night's backup now!) much easier. Backups are handy for recreating test copies of databases too. I've posted the prune compress script here before, so it should be in the archives somewhere. It's nothing fancy. -- Russell Howe, IT Manager. [EMAIL PROTECTED] BMT Marine Offshore Surveys Ltd. - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Database odbc or one more driver.
Chris Hoogendyk wrote, sometime around 20/08/07 21:24: my gut reaction was ... I suppose you could write such an interface, but why? If someone insists on a commercial database, why would they not insist on a commercial backup solution? Or, alternatively, if someone was willing to accept an open source backup solution, then what's the big deal about doing a simple install of mysql? Seems like the audience for bacula using DB2 would be miniscule. One reason might be that if you've shelled out megabucks on a commercial database, you've probably also bought some decent hardware to run it on. Setting up a second database server when you've gone to the trouble to configure one for decent performance is quite a bit of work (not to mention somewhat costly). I'd love to get working on SQL Server support for bacula - then we could drop our postgres server and free up quite a bit of capacity on one of our Xen hosts. We're soon to set up a clustered install of SQL Server and once that's in, and things quieten down, maybe I'll finally get chance to look at it. SQL Server isn't going away any time soon (at least a couple of critical apps here are written to work with SQL Server and SQL Server alone - even to the point of using encrypted stored procedures, no less!). ODBC would seem like the way to go, although it could just use the TDS libraries directly I suppose.. -- Russell Howe, IT Manager. [EMAIL PROTECTED] BMT Marine Offshore Surveys Ltd. - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] bacula 2.0.3 for Win32 can't backup top level Chinese folder!
Kern Sibbald wrote: If anyone on this list knows of a Win32 tool that will write UTF-8 (not Unicode or UTF-16), please let me know. It's not really a win32 tool, but jEdit (text editor, written in Java) can output in pretty much any character set. http://www.jedit.org/ See Buffer Options for the character set selector (you can also switch between different line terminators). I think even Windows notepad is unicode-aware these days, although whether it uses UTF8 or one of the other encodings, I don't know. -- Russell Howe [EMAIL PROTECTED] - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Coping with brief network outages when talking to clients
Morning all, I'm not sure if this has been discussed before, but I'm looking at running the Windows version of the FD within the Windows 2000 cluster service. This looks to be possible, with the cluster service handling the stopping starting of the process and if the configuration is stored on a drive which the cluster has control of, then the config can fail over with the process. Windows lists the requirements for a cluster-unaware application to be clustered as follows: * The application must be able to store its data in a configurable location. Check - config file is a command line parameter The application must use TCP/IP to connect with clients. Check Clients must reconnect in the event of intermittent network failure. This might be a problem - in my experience, a reset of the TCP connection will cause the job to terminate. There is an option to rerun failed jobs, but it would be nicer if the director and sd could hold off terminating the job for a minute or so whilst they try to re-establish the connection. If a job is re-run, presumably it is re-run from the beginning, so if we were doing a backup of 190GiB of data onto a 200GiB tape, and it got to 180GiB and then the cluster node went down, when bacula reruns the job and connects to the fd again (now running on the backup cluster node) presumably it will begin the job from the start and overflow the tape (although I suppose it could rerun it as a differential...). Has this ever been considered, and how much work would it be to implement? -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] alternative postgres config
Hans Manz wrote, sometime around 14/02/07 20:36: Hello, folks! I've found an alternative way to configure postgres than in the doc. I've installed an identd on my postgrest/bacula-dir server. In pg_hba.conf I wrote the line: local bacula bacula ident bacula I don't think you need identd for this - when using unix sockets, postgres gets the other user's uid via a syscall on the socket, IIRC. -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Hierarchical storage systems?
Alan Brown wrote, sometime around 14/02/07 12:52: I know this is outside the scope of the list, (despite someone having this on the wishlist), but I'm looking for a *nix-compatible (pref linux) hierarchical storage system. I don't know if it's of much help, but XFS has an API for Hierarchical Storage Management (HSM) apps called DMAPI. It's an SGI thing, and there must be at least one or two HSM apps out there. -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] WOW
Janco van der Merwe wrote, sometime around 11/09/06 15:23: Yes but now you have to have a 3rd party app and requires subscription but Vodacom still has the e-mail to SMS but that is limited to 6 a day which will work but we are currently using the MTN network with our Zabbix which is on another machine and I hoped that there was a possibility that I could some how configure Bacula to use Zabbix which will solve this but I mean this is not a necessary thing just a nice to have, at least when jobs fail I have prior notice so that when my boss corners me I at least know and give him that all wonderful IT answer of I know and I'm busy working on it You'll be needing a commandline application which can send an SMS, given some text to send, and a number to send it to (this could be hardcoded into the application). Something like ./sendsms --to +4477123123123 This is an SMS Once you have that, it's easy enough to interface to Bacula, in the same way that job results are emailed. You'll probably find the same tool useful for a bazillion other things (pushing your bill up in the process :) -- Russell Howe [EMAIL PROTECTED] - 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=lnkkid=120709bid=263057dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] WOW
Alan Brown wrote, sometime around 11/09/06 16:05: Personally: I have my phone on an unlimited data tariff I wasn't aware such a thing existed in the UK, and certainly isn't all that common around the world. A friend of mine was charged about £200 for 6MiB when living in Canada recently! Admittedly, he probably wasn't on the best of tariffs, but still.. £200! and just use a Symbian imap client to check mail every 15 minutes No message length limits, no overly long delays (SMSes usually deliver within seconds, but I have seen even local network ones take 6 - 24+ hours to deliver, depending on network load) and no stratospheric SMS charges (10c/message * 160 bytes maximum = $625/MB!!!) Granted, this doesn't work with el-cheapo handsets, but many phones have some form of pop3/imap client onboard these days. Yup - I got a new one this Saturday, and it does POP3, IMAP and SMTP, all with SSL as an option (it even validates the certificate!). Most impressed, I was, until I saw what Orange wanted to charge me for GPRS on my tariff :) -- Russell Howe [EMAIL PROTECTED] - 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=lnkkid=120709bid=263057dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] FileList scripting help
Davis, Jacob wrote, sometime around 09/08/06 14:16: It appears that a crucial line break was missing from my original post. To be clear, get_newest_file.py accepts a single path argument and returns the fully qualified path to a single filename like so: $ get_newest_file.py /usr/local/backups/ /usr/local/backups/newfile.tar Any ideas why Bacula does not like my FileList? File = '|get_newest_file.py /usr/local/backups/' Could not stat '|get_newest_file.py /usr/local/backups/': ERR=No such file or directory Bacula is looking for a file called: |get_newest_file.py /usr/local/backups/ Take out your single quotes -- Russell Howe [EMAIL PROTECTED] - 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=lnkkid=120709bid=263057dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Estimate + Exclusion Question
Kern Sibbald wrote, sometime around 03/08/06 20:56: On Thursday 03 August 2006 20:45, Erich Prinz wrote: 2. How do we determine if the exclusions are working? Try the following: estimate job=job-name listing level=Full Wow, Kern I love you! If I'd known about that (presumably if I'd read the documentation a bit more carefully), I could have saved many many an hour of messing around before. -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] restore to local file
Tom Newton wrote, sometime around 03/08/06 12:10: I have a machine that is no longer with us. I wish to restore one or two files from the backup set, but obviously, without that machine, I am stumped - bacula wants to restore to the client it read from, i'd rather it gave me the files in /tmp on my server. Can I do this? How? Choose the 'mod' option when running a restore job and alter the client. That one's gotta be in the manual, surely? :) -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] restore to local file
Tom Newton wrote, sometime around 03/08/06 12:44: On Thu, 2006-08-03 at 12:15 +0100, Russell Howe wrote: Choose the 'mod' option when running a restore job and alter the client. That one's gotta be in the manual, surely? :) thanks, did not read what that one's for. Tried it, but on my debian system it says my file daemon won't support win32 streams. On remote windows systems I get Aha, another gem hidden in the documentation. Unless you specify that Windows backups are written in 'portable' format, the Windows backup API is used which writes data in its own 'special' (for 'special' read 'proprietary'). Bacula doesn't (yet, but it's on the todo) parse this data, just storing it and returning it to the Windows backup API for decoding. This effectively means that you need to restore to a Windows machine, or you need to perform a backup of your ex-machine with the portable option set. See the docs for more information: http://www.bacula.org/rel-manual/Windows_Version_Bacula.html#SECTION000396000 I doubt you'd be able to run the win32 FD under wine as a way around this, but the wine folks do implement all sorts of esoteric parts of the Windows API so anything is possible, I guess.. -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] restore to local file
Tom Newton wrote, sometime around 03/08/06 15:05: On Thu, 2006-08-03 at 14:57 +0100, Russell Howe wrote: It should be a path which exists on the target client. Tried C:/tmp which I *know* exists, but still getting the same errors. Starting to really get me down this one. From your previous message - Socket error from Filed on Include command: ERR=Connection reset by peer It said the fd closed the connection - anything in the FD's logs (on the Windows box) explaining why? Do you have any kind of firewall in the way? (including the built in Windows XP firewall!) - looking at your email address, one would assume you knew to check for packet filters ;) -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] restore to local file
Tom Newton wrote, sometime around 03/08/06 14:44: Any idea why it fails to back up to another windows XP box? effectively Do you mean restore (files) to instead of back up to? one of its siblings... should I change the path from /tmp? It should be a path which exists on the target client. e.g. I have a 'restored files' on a drive lettered T:, so I restore to: T:/restored files HTH -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] restore to local file
Tom Newton wrote, sometime around 03/08/06 15:19: On Thu, 2006-08-03 at 15:11 +0100, Russell Howe wrote: It said the fd closed the connection - anything in the FD's logs (on the Windows box) explaining why? Where might I find such a log? Hmmm.. I thought it kept one, although I fail to see where now that I'm actually logged into one of our Windows boxes. Is there anything which looks promising on the right click menu for the filedaemon's tray icon? I'd run tcpdump on the intermediary firewall and see if you spot the TCP connection being closed - that will at least potentially help you to debug this.. Also, look at netstat -ano output and match the pid of bacula-fd.exe (get it from Task Manager - you'll have to add in the PID column using 'View' - 'Columns' or somesuch) to see which ports it is listening on. Once you know the port numbers, you should be able to telnet to them from the director and the fd and get a connection (along with warning messages from the fd about weird connections :) -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] restore to local file
Kern Sibbald wrote, sometime around 03/08/06 13:54: On Thursday 03 August 2006 13:56, Russell Howe wrote: Unless you specify that Windows backups are written in 'portable' format, the Windows backup API is used which writes data in its own 'special' (for 'special' read 'proprietary'). Bacula doesn't (yet, but it's on the todo) parse this data Parsing this data *is* implemented in 1.39.18. ... and bacula continues to astound! Thanks all who contribute -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] restore to local file
Tom Newton wrote, sometime around 03/08/06 16:04: Actually.. there is a reset comes from the FD - the last packet with any data contains: ..2000 O K Job Wi 0050 6e 64 6f 77 73 20 58 50 2c 4d 56 53 2c 4e 54 20 ndows XP ,MVS,NT anyone want to see a dump? Hm, not me :) Still, it looks like the FD is closing the connection for some reason - are you running the exact same versions of bacula on director, file daemon and storage daemon? If you're not, things can get 'interesting'... -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] restore to local file
Tom Newton wrote, sometime around 03/08/06 16:13: On Thu, 2006-08-03 at 16:07 +0100, Russell Howe wrote: are you running the exact same versions of bacula on director, file daemon and storage daemon? If you're not, things can get 'interesting'... Ah! You may be onto something! The version of bacula I have is debian sarge 1.36.2-2sarge1 (ie about 30 years old) but the windows client is 1.38.4 - are my backups likely to be worthless? If I install an older fd will that work, or should I compile a new director from source? Try an older fd on Windows (replacing the .exe file whilst the service is stopped should be sufficient - I don't think the configuration format has changed much). Stop the bacula service, take a copy of the current bacula-fd.exe program and extract a 1.36.2 (or .3) bacula-fd.exe from the archive which presumably is up on sourceforge somewhere and then start the bacula service. See if you can restore then. I would expect your backups to be restorable, but well, you never know until you try :) -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Feature request - SQL Server (freetds? odbc?) support for catalog database
I didn't see a feature request on the site for extra database support, so here goes: Origin: Russell Howe [EMAIL PROTECTED] Date: 26 July 2006 What: It would be nice for the database backend to support more databases. I'm thinking of SQL Server at the moment, but I guess Oracle, DB2, MaxDB, etc are all candidates. SQL Server would presumably be implemented using FreeTDS or maybe an ODBC library? Why:We only really have one database server, which is MS SQL Server 2000. Maintaining a second one for the backup software (we grew out of SQLite, which I liked, but which didn't work so well with our database size). We don't really have a machine with the resources to run postgres, and would rather only maintain a single DBMS. We're stuck with SQL Server because pretty much all the company's custom applications (written by consultants) are locked into SQL Server 2000. I can imagine this scenario is fairly common, and it would be nice to use the existing properly specced database server for storing Bacula's catalog, rather than having to run a second DBMS. I haven't seen Bacula's database routines, but it seems like it's fairly modular and so I'm guessing that adding another database wouldn't be too much work. What would be involved? -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Feature request - SQL Server (freetds? odbc?) support for catalog database
Dan Langille wrote, sometime around 27/07/06 16:29: On 26 Jul 2006 at 11:55, Russell Howe wrote: I haven't seen Bacula's database routines, but it seems like it's fairly modular and so I'm guessing that adding another database wouldn't be too much work. What would be involved? I wrote the PostgreSQL module. Things get added because the person adding it *wants* to add it. The result: SQL Server will not get added until someone codes it. I'm not interested in SQL Server support, so it will not be me. Of course - I should perhaps have sent the email as an I'd like SQL Server support for the catalog - anyone interested in helping me do it? email instead of a feature request. This is the way it is. Nothing personal against anyone. It's just how things work. Oh I wasn't expecting others to really do it, but since Kern wanted all feature requests submitted as per his template, and I want SQL Server support, I figured I'd submit it. It's something I plan to do, but I haven't had time to do anything but fill in the feature request form. I was more after feedback on how to go about adding support for another database and how much trouble was involved. It would take knowledge of C coding, knowledge of SQL Server, and good debugging skills. The C I can handle, the SQL Server I can probably figure out (with the help of freetds etc) and the debugging is mostly a matter of dedicating time to it. I'd love to contribute something back to bacula, and as soon as the opportunity presents itself, I will. -- Russell Howe [EMAIL PROTECTED] - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.phpp=sourceforgeCID=DEVDEV ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Problem with the list???
Robert W Hartzell wrote, sometime around 23/05/06 22:06: Has anyone else noticed a problem with delivery of the list messages? I seem to be getting messages today that are 4 or 5 days old and 2 messages that I sent never showed up on the list. That'll be me, then... I haven't been going through the held postings daily, as I ought to be. Essentially, if you want your postings to show up on the list in a timely fashion, it's best to subscribe and send from your subscribed address. Otherwise, you'll be relying on me or Kern to let your postings through. If you don't want to be deluged with messages, you can subscribe and set yourself to nomail. As to the two which got lost, I guess it's possible I discarded them by accident, but I try to be careful.. It could be possible that Sourceforge is having problems, although I haven't seen anything to suggest that. Your friendly local list admin, -- Russell Howe [EMAIL PROTECTED] --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Problem with the list???
Martin Simmons wrote, sometime around 24/05/06 15:56: Looking at the Received: headers might show where the delay occurs. FWIW, the delayed messages that I see are usually duplicates, possibly because they are resent by the poster on the original date after subscribing and then the original is flushed from the held queue later. Back in those heady days when I could actually find time to read all the postings to the list, I'd notice these duplicates and discard them from the held messages queue. As it is, I just let everything that's a legitimate message through, as I can't find time to keep up with everyone's messages on the list (this is a good thing - a nice healthy, active user community!). -- Russell Howe [EMAIL PROTECTED] --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Backups too big, and other questions
Eric Warnke wrote, sometime around 27/04/06 03:42: Unfortunately I have looked high and low, there is just no good way to read sparse files intelligently. Whoever though of providing the functionality without an API to step through the block mapping was a moron. It is truly a brain dead technology. At least there is a fcntl under Win32 to deal with it with some intelligence. As far as I know, there is no cross-platform way to do it, and in Linux, no filesystem-agnostic way to do it. XFS has a special ioctl, XFS_BMAP, or something like that. It returns a list of extents occupied by the file, and from there you can work out where the gaps are. I think there were whispers on the linux-xfs/l-k mailing lists about genericising this across all file systems, but I don't know if it actually got taken up as a serious proposition. Anyway, at least in XFS, it's possible. # dd if=/dev/zero seek=50 of=foo count=100 100+0 records in 100+0 records out 51200 bytes (51 kB) copied, 0.000916 seconds, 55.9 MB/s # xfs_bmap foo foo: 0: [0..47]: hole 1: [48..151]: 20688..20791 # Here we can see that the first 48 blocks are hole, and the next 103 blocks contain data (in this case, zeros). Presumably bacula would have read this entire file as being one big hole, had it been backed up with sparse turned on? If so, this could lead to a substantial disk space discrepancy on a system if you back up with sparse turned on, and then restore. (you'll have 'gained' any disk space previously occupied by zeros). And no, you don't need to be root to perform the ioctl. -- Russell Howe [EMAIL PROTECTED] --- 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=lnkkid=120709bid=263057dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] backing up msexchange or mssql
James Harper wrote, sometime around 29/04/06 06:11: Is anyone doing this now? Would there be any value in developing an fd module specifically to back up these databases? I do backups of MS SQL 2000 to files and then gzip them (reduces the backups to about 10% of their original size). Doing this, I keep about a week's worth of backups on the disk, which are all backed up every night. It's worked well for over a year now. I have this little shell script to handle gzipping the files and pruning the them when they get too old. I wrote it to use Cygwin, but it would probably be better written in perl or python etc. -- Russell Howe [EMAIL PROTECTED] backupmgr.sh Description: application/shellscript
Re: [Bacula-users] FOSDEM 2006
Jo wrote: Russell Howe schreef: Is anyone else going to FOSDEM 2006 in Brussels? I'm living 30 kilometers from Brussels. I'll probably be there on Sunday and maybe Saturday afternoon. My cell phone number is +32 (0)478-33 76 27. Maybe we'll meet there, Oh, sorry - I just received your email now! Hope you enjoyed the conference - I certainly did. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnkkid=110944bid=241720dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] A script to ease the pain of switching from SQLite to PostgreSQL
Kern Sibbald wrote: On Monday 06 February 2006 18:04, Russell Howe wrote: Well, the time is nigh. Our SQLite database has hit 1.3GB and creates a 900MB dump file, which is big enough to cause our bacula Xen virtual machine to run out of space when running a catalog backup. We have a Postgres (7.4) server with plenty of space, so I thought I'd try importing the catalog into that.. things didn't go so smoothly, so I ended up knocking up a script to do the import. I haven't actually tested this yet, so it's probably broken and riddled with bugs. Still, enjoy :) I've added this to the examples/dbdump directory -- perhaps not the most obvious place. If you ever have an update, please send it to me. Thanks for the contribution, in addition to your regular Bacula list spam filtering work :-) heh, I should add that the script is pretty broken. I've got a more up to date version at http://siksai.co.uk/~rhowe/software/sqlite2pgsql but it's still not quite right. Still, it may make the transition slightly less painful for anyone else who has to do it. In the end, I blew our catalog away after getting complaints from SQLite about the database file being corrupt. I started afresh with Postgres. We were switching to a new drive and a whole new set of tapes, so the catalog was only of historical interest in case we needed to restore from a while back. In about 4 months, it would be rendered completely useless. I'm tempted to knock up yet another web interface to the database, however, so you might see some Java stuff from me in the not too distant future. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnkkid=110944bid=241720dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] FOSDEM 2006
Is anyone else going to FOSDEM 2006 in Brussels? http://www.fosdem.org/ I'll be arriving Friday afternoon, and will stay until sometime Sunday afternoon/evening. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnkkid=110944bid=241720dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: JobSet, Batch, or Aggregate Concept / Max tape usage per (WAS: Re: [Bacula-users] Hit me w/ a Clue-by-Four (Amanda user))
Brian A. Seklecki wrote:
On Thu, 2006-02-09 at 10:37 +, Russell Howe wrote:
Bacula will do this. Check you don't have Maximum Volume Jobs = 1 in
I do. This is because I'm trying to enforce a specific threshold of a
single-tape-per-day policy.
Ah, but this is a single-tape-per-job policy.
I can manage volume capacity myself. I'd like to override Bacula's
default behavior of auto-picking a tape based on status/capacity/recycle
time.
I'd essentially like to Micro-Manage it perhaps.
I want my tapes to be written to by the scheduled jobs for their daily
assignment, by all the jobs scheduled to run at that time, and then be
marked Used.
The trick is, if all the jobs run at the same time, don't mark the
volume Used or Full until after they've all run.
When it comes to marking tapes as Used or Full, there seem to be two
ways to do it currently:
* Maximum Volume Jobs
* Volume Use Duration
So, you can either say 23 hours after the first job which wrote to this
tape {started,finished}, mark it as used to get you a new tape every
day (I'm not sure which of those two options applies, but I guess it's
in TFM somewhere)
Or, you can say After n jobs have been written to this tape, close it
off/mark it used
The latter is what I do. I run 10 jobs every night, most of them on the
same schedule, so they start at the same time, although they don't run
concurrently - I have Maximum Concurrent Jobs = 1 set for the storage
device, so that jobs get written sequentially.
After the last job completes (the Catalog backup), I get this:
11-Feb 21:43 bacula-dir: Start Backup JobId 26,
Job=BackupCatalog.2006-02-10_23.51.00
11-Feb 21:43 spanky-sd: Volume PoolA_Weekly_1 previously written,
moving to end of data.
11-Feb 21:47 spanky-sd: Ready to append to end of Volume
PoolA_Weekly_1 at file=95.
11-Feb 21:47 bacula-dir: Max Volume jobs exceeded. Marking Volume
PoolA_Weekly_1 as Used.
This does mean that if I change the number of jobs which get written, I
have to remember to update the catalog's record for the media, and
update the Pool definitions in the director's configuration file. I have
this comment in the configuration to remind anyone who might inherit
this setup:
If you're not seeing the above behaviour, I guess there could be some
kind of race condition whereby when two jobs run concurrently, and both
try to incrememnt the number of jobs field in the catalog for the
media table, both jobs update it to n+1 instead of the first updating it
to n+1 and the second adding one to that (making it n+2)... I find it
hard to believe that such a bug would go unnoticed, though... maybe
nobody uses Maximum Volume Jobs with concurrent backups?!
--
Russell Howe
[EMAIL PROTECTED]
---
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: JobSet, Batch, or Aggregate Concept / Max tape usage per (WAS: Re: [Bacula-users] Hit me w/ a Clue-by-Four (Amanda user))
Russell Howe wrote: I have this comment in the configuration to remind anyone who might inherit this setup: # ** VERY IMPORTANT ** # In the database, the media entries have a MaxVolJobs setting. This is # what determines that a tape is full. # # Current values are: # Cases tapes: 10 (Cases, Cases2, Artemis, Zetafax, Thor, Zeus, Users # Depts, Bankside, SQL Server, Catalog) # # This is changed much like the above, i.e. # UPDATE media SET MaxVolJobs = 1 WHERE VolumeName LIKE 'PoolA%'; -- Russell Howe [EMAIL PROTECTED] --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: Date/Time poll'ing ? (WAS: Re: [Bacula-users] Simulate/Test Scheduler?)
Martin Simmons wrote: On Sun, 12 Feb 2006 16:18:01 -0500, Brian A. Seklecki [EMAIL PROTECTED] said: How often does the Director poll the system date time? How often does the Storage director? I don't see any options in the config to control this. Specifically, for periodic maintenance like recycling/pruning/volume status flagging. AFAIK, these maintenance operations are never run on their own. They are always triggered by other operations such as a job running or 'status dir' when Bacula needs to decide which volume to use. You can trigger them by running a job of type Admin -- Russell Howe [EMAIL PROTECTED] --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] user complains bacula slows his laptop
Alex Finch wrote: Dan and Michel, Thanks for the helpful comments and suggestions. Next time the user is here I will try some of them out! I believe he is running Windows XP professional, not sure of the exact spec. of the machine but I imagine it is fairly good. One other thing he can try (assuming he has permission to do so - he might not) is lowering the process priority of the FD when it's running. You can do this from Windows Task Manager, although if it's running under the local SYSTEM account, you might not be able to change it, even as Administrator... -- Russell Howe [EMAIL PROTECTED] --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] A script to ease the pain of switching from SQLite to PostgreSQL
Well, the time is nigh. Our SQLite database has hit 1.3GB and creates a
900MB dump file, which is big enough to cause our bacula Xen virtual
machine to run out of space when running a catalog backup.
We have a Postgres (7.4) server with plenty of space, so I thought I'd
try importing the catalog into that.. things didn't go so smoothly, so I
ended up knocking up a script to do the import.
I haven't actually finished it yet, and I know there are some
optimisations which could be made, but I hope this is useful to someone:
What could be better:
* Automatic reorganisation of CREATE statements so that they execute in
the correct order
* Commandline parameters
* Handling of arbitrary conversions (Postgres - MySQL, MySQL -
Postgres, SQLite - MySQL, MySQL - SQLite, Postgres - SQLite)
Note that it would be a good idea to turn off the fsync stuff on
Postgres while you do the import - I haven't benchmarked, but it should
make things go a good bit quicker.
I haven't actually tested this yet, so it's probably broken and riddled
with bugs. Still, enjoy :)
--
Russell Howe
[EMAIL PROTECTED]
#!/bin/bash
# Import an SQLite dump of a Bacula catalog into Postgres
# Designed for v1.63.3 (as found on Debian sarge)
#
# v0.5
#
# Copyright (c) 2006 Russell Howe [EMAIL PROTECTED]
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the Software),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY KIND, EXPRESS
# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
FILE=bacula.sql
# Tables, in order of size
TABLES=(File Filename Path Job Media Pool CDImages Counters Version Client
FileSet JobMedia NextId UnsavedFiles BaseFiles)
# Tables, in insert order
TABLESINSERT=(Pool CDImages Client Counters FileSet Filename Job Media JobMedia
NextId Path File UnsavedFiles Version BaseFiles)
DBNAME=bacula
LOGFILE=/var/tmp/sqlite2pgsql.$$.log
importdata() {
if [ x == x$1 ]; then
echo Error: importdata() called without an argument. Aborting.
exit 1
fi
SQLFILE=$1
if [ ! -r $SQLFILE ]; then
echo Error: Cannot read from $SQLFILE. Aborting.
exit 1
fi
echo -n Loading $SQLFILE into database $DBNAME...
psql -d $DBNAME -f $SQLFILE || (
echo Failed to load $SQLFILE into database $DBNAME. psql
exited with return code $?. Aborting.
exit 1
)
}
# Go through each of the table names, splitting the INSERT statements off
# into seperate files
for table in [EMAIL PROTECTED]; do
SRC=$FILE.other
if [ ! -f $FILE.other ]; then
SRC=$FILE
fi
PATTERN=^INSERT INTO $table
if [ ! -f $FILE.data.$table ]; then
echo -n Separating $table table from database dump...
echo BEGIN TRANSACTION $FILE.data.$table.tmp
grep $PATTERN $SRC $FILE.data.$table.tmp
echo COMMIT $FILE.data.$table.tmp
mv $FILE.data.$table.tmp $FILE.data.$table
echo done. ($FILE.data.$table)
echo -n Stripping matched lines from the source file to speed
up the next round...
grep -v $PATTERN $SRC $FILE.other.tmp
mv $FILE.other.tmp $FILE.other
echo done.
else
echo $FILE.data.$table already exists. Assuming this table has
already been split
echo off from the main dump. Not regenerating.
fi
done
echo Seperating DDL statements from INSERT statements
grep -v ^INSERT $FILE.other $FILE.ddl
echo DDL statements are now in $FILE.ddl
grep ^INSERT $FILE.other $FILE.data.other
echo Any remaining INSERT statements are now in $FILE.data.other
echo Fixing up datatypes used in the DDL...
sed -e 's/TINYINT/SMALLINT/g' \
-e 's/DATETIME/TIMESTAMP/g' \
-e 's/INTEGER UNSIGNED/INTEGER/g' \
-e 's/BIGINT UNSIGNED/BIGINT/g' \
-e 's/INTEGER AUTOINCREMENT/SERIAL/g' \
-e s/\ DEFAULT\ \\/\ DEFAULT\ \'\'/g \
-e s#\ TIMESTAMP\ DEFAULT\ 0#\ TIMESTAMP\ DEFAULT\ \'1/1/1970\'#g
$FILE.ddl
Re: [Bacula-users] Please remove me from the list
If you're really stuck with usage of the list, email the list admin at [EMAIL PROTECTED] Don't bother the list with administrivia. -- Russell Howe, bacula-users moderator [EMAIL PROTECTED] --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://sel.as-us.falkag.net/sel?cmd=lnkkid=103432bid=230486dat=121642 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Don't run as root if you can avoid it.
James P. Kinney III wrote: bacula is a member of the disk group. Isn't being a member of the disk group effectively the same as being root, since you can gain access to the block devices containing the local filesystems? -- Russell Howe [EMAIL PROTECTED] --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] DLT tape errors suddenly start
On Tue, Jan 10, 2006 at 07:59:05PM -0500, Dan Langille wrote: My DLT tape drive has been going along nicely for almost a year. Tonight, the day before I head away for a trip, it starts giving errors. You did, presumably, try a cleaning tape in it? :) -- Russell Howe | Why be just another cog in the machine, [EMAIL PROTECTED] | when you can be the spanner in the works? --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] single instance storage?
Mordechai T. Abzug wrote: Does bacula support storing only one instance of a file that is identical accross multiple machines? Ie. if the contents of, say, /bin/ls are the same on a bunch of different hosts, is bacula smart enough to only store it once on the volume? If you look at the recent Project voting email, I think this is what Kern terms Base Jobs. In short, no, not yet, but it's planned I think. If you DO support such features, does bacula also support some sort of logarithmic repitition for popular instances? Ie. if the file is present on, say, 3 different systems, I might only want one copy of it. But if it's on 50 different systems, I might want a few copies of it, so if one of them has a media error, I don't lose this (very important) file. Do you have this feature, too? Might be a nice tweak on the implementation of 'base jobs', although the restore code would need tweaking to handle the fact that data for a file could be found in multiple places. Also, is bacula hard link aware? Ie. if I have a whole bunch of large files with a bunch of hard links to them, and I restore, is the hard link structure preserved? As far as I know, yes. I'd check with the manual on this (I think there used to be performance issues with large numbers of hard links or something - not sure). What I presume you're thinking of doing is using backuppc (or whatever the hardlinking backup software thingie is called) to provide a hardlinked backup of your boxes, and then back that up with bacula to save space. I think that should be possible in the absence of base jobs. -- Russell Howe [EMAIL PROTECTED] --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] single instance storage?
Mordechai T. Abzug wrote: On Thu, Dec 08, 2005 at 12:05:40PM +, Russell Howe wrote: If you look at the recent Project voting email, I think this is what Kern terms Base Jobs. In short, no, not yet, but it's planned I think. Ah, thanks. I've only just joined the list, but the archives have it. Base jobs sounds similar, but not quite the same -- the intent of this is to catch not only OS and apps, but even data files that are identical (ie. a bunch of users have a copy of the same 50MB powerpoint presentation), and it (should be) automatic. If things were implemented the way you describe below, then it would seem that base jobs would be possible as a specific case, with no code changes needed. Ie. after archiving any file, you store the file size and the crypto checksum in a DB, right? If you later encounter another file with the same size and crypto checksum, instead of backing it up, you put a reference in your DB that says it's the same content as the earlier file. Restores may take longer, because files are spread out, but if the restore software is smart, it should be able to reorder its queries so it only needs to make a single pass over the backup media. I think this would be a lovely feature to have, but it's not the way bacula works at the moment, and it would probably require quite some code changes, and changes to the way restore behaves. I guess care needs to be taken here - restores are not something you want to break :) I've been using AMANDA for a while, particularly because it can easily support bare-metal-recovery and it's free. But I'd like something that supports single instance storage, and has better support for multiple volumes per backup. Guess bacula isn't it, at least not yet. :( No, although you could possibly emulate it by sticking whatever backupPC commands you need to use as a ClientRunBeforeJob and then backing up the backupPC tree, although restores would be irksome. The multiple volumes per backup bit is quite a 'selling point' for bacula in the open source backup world at the moment I think. -- Russell Howe [EMAIL PROTECTED] --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: FW: [Bacula-users] Performance testing and tuning
Ribi Roland wrote: With spooling it takes arround 10-20min to spool data and write them down to tape at 2MB/s. But the save of the attributes (writing to the catalog) needs 2h or more! Looks like the catalog database is underperforming, although maybe 2h is reasonable if you have millions of tiny files (which will be relatively quick to back up, but will take a long time to record in the catalog). I tried many combinations and recompiled bacula also with Postgres 8.1, Postgres 8.0, with CFLAGS=-O2/CXXFLAGS=-O2 I compiled it with gcc 3.4.4, which I compiled myself, because on solaris 10 there is'nt any binary distribution. I would recompile as a last resort, not as a first one... I think that there are 2 Problems, one is the performance of Bacula or Postgres itself Postgres, by default, does an fsync (or fdatasync, can't remember) after every write operation. This can kill performance, especially when you don't use transactions (bacula doesn't). Since the bacula database is backed up, and since it can be recreated with bscan, I think I would be happy turning the fsync behaviour off and leaving my database open to corruption in the event of a system crash or power loss. This information, and a lot more, can be found in the Postgresql performance tuning FAQ, linked to from the Postgresql website. and also some problems with the tape. I did'nt get more then 2.6Mb/s with the ufsdump from solaris. Are the only devices on the SCSI bus the drive and the controller? Not knowing anything about Solaris, I don't know what you can do with the OS to query/test the bus performance. -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Error with win32 fd caused director to block, awaiting manual intervention on the client
After seeing people post about performance improvements with a larger
network buffer size, I thought I'd give it a try on one of our Windows
file daemons.
My attempt failed, but more worryingly than the configuration change not
having the desired effect (this may be due to lack of RTFMing on my
part), the failure of the Windows file daemon caused all further jobs on
the same storage device to block, waiting for me to hit 'OK' on the
console of the Windows server.
The director was showing this:
1920 FullArtemis.2005-11-04_22.05.05 is waiting on max Storage jobs
1919 FullZetafax.2005-11-04_22.05.04 is waiting on max Storage jobs
1918 FullThor.2005-11-04_22.05.03 has terminated
Thor being the job which had errored.
Presumably thor-fd was still holding the connection to the sd open, and
so Zetafax couldn't get started.
The message the fd was logging was this, from bacula/src/lib/bnet.c:
if (dbuf_size % TAPE_BSIZE != 0) {
Qmsg1(bs-jcr, M_ABORT, 0,
_(Network buffer size %d not multiple of tape block
size.\n),
dbuf_size);
}
Could the win32 fd not log this message either to the director, or to
the Windows Event Log instead of popping up a dialog box and waiting for
operator intervention?
Luckily, the machine was local, and intervening wasn't difficult, but
had it been remote, things could have been trickier.
It worries me that a failure on a client can affect jobs queued for
other clients in this way.
--
Russell Howe
[EMAIL PROTECTED]
---
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. Download
it for free - -and be entered to win a 42 plasma tv or your very own
Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] mail notifications
Kern Sibbald wrote: On Monday 31 October 2005 15:38, Viktorija Almazova wrote: He will be shocked and every day will ask email description. So if is there a way to make these messages like this: Hello, Dear user! This is possible if you intercept the emails, process them, and send your own, or if you write a Python script that checks the job status and knows where to send the email -- of course, this would require 1.38.0 Can't he just set a mailcommand which takes the necessary information on the commandline and which calls bsmtp/sendmail/mailx/ssmtp/etc with the appropriate options? -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Re: Database Must Be Broken - Was Restore Doesn't List Any Files?
On Thu, Oct 27, 2005 at 09:49:11AM -0700, Drew Tomlinson wrote: On 10/27/2005 8:47 AM Drew Tomlinson wrote: A little more poking around and I find that basically nothing works in bconsole so I suspect my database is extremely corrupted. So now I'm off to Google in search of the recommended method to wipe out the current corrupted database and restore from my nightly catalog backups. However if someone could nudge me in the proper direction, I'd really appreciate it. You don't say which database you're using, which affects the specifics of the restore process somewhat! I think there's a dbcheck program which is part of bacula, which will check your database for consistency, although someone did mention that with a large catalog it can eat an extraordinary amount of RAM (kinda like xfs_repair does on a large XFS filesystem, I guess). I think the procedure would be something along the lines of: * Torch your existing database (or rename it) * Run make_bacula_$dbname_tables (or whatever the script is called) to create a fresh, blank, database. You could create it under a different name to the corrupt one if you can't rename an existing database... * Make sure your director configuration is the same as before (not sure if this is necessary, but it can't hurt), with only the name of the database changed if necessary * Use bscan to repopulate the database so that it holds information about the tape containing your latest catalog backup. * Restore the catalog backup (probably a single text file containing SQL statements to repopulate the database) * Import the database dump, checking to see what it's going to import into, and that it isn't going to clobber anything important * Hope that the catalog backup you restored wasn't corrupted. You may be able to use bextract instead of bscan+restore to extract just the database dump, which would probably save you considerable time and fustration. -- Russell Howe | Why be just another cog in the machine, [EMAIL PROTECTED] | when you can be the spanner in the works? --- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] *Extreme* low MySQL performance when prunnig or purging jobs/volumes
Florian Schnabel wrote: i'm using postgresql now ... database cleanup usually takes 5-10 minutes, CPU less that 1, seems to be only limited by HD speed ... Check to see if Postgres is issuing an fsync() after every operation or not, and then decide if you need it to. If you can disable that, things should get a *LOT* faster. I posted a link to some Postgresql performance tuning documents a while back - they're linked to from the postgres website too. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] dev open failed
Arno Lehmann wrote: Hello, sourceforge holds mail again? Anyway, I've just got a bunch of mail dated 21 October... List moderator had an oh crap moment and had a clearout of the posts held for moderation (there were over 40 scam emails in there, as well as a bunch of held posts). -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] BackupCatalogue errors
Arno Lehmann wrote: Hello, On 26.10.2005 15:20, Chris Boyd wrote: hq-dir: message.c:462 Mail prog: bsmtp: bsmtp.c:276 Fatal connect error to localhost: ERR=Connection refused This looks like either a permissions problem or a full disk to me. Looks more like there isn't a mail server running on the host, or if there is, it's not listening on 127.0.0.1 I assume the error is coming from the host running the director (neither the sd nor the fd seem to have the ability to send email alerts). -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] bacula-fd service doesn't respond
Kern Sibbald wrote: On Wednesday 26 October 2005 17:15, Shawn Wilson wrote: Has anyone been able to get the bacula-fd service to work on a win2k3 server? I ran the default install routine and had it install itself as a service but when I run 'netstat -an', nothing is listening on port 9102. Have you checked to see if the installed service is being started automatically by Windows? Check the 'Services' MMC snapin. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] *Extreme* low MySQL performance when prunnig or purging jobs/volumes
Paul Heinlein wrote: At the risk of being somewhat off-topic, I'll add that the same is true of SQLite. In SQLite 2.8, setting the default_synchronous flag to OFF will speed things up markedly: echo 'PRAGMA default_synchronous=0;' | sqlite bacula.db I posted about this previously: http://article.gmane.org/gmane.comp.bacula.user/16853/match=pragma Looks like they just removed the option to default all databases to non-synchronous operation in SQLite 3.0, requiring you to manually set your databases to be asynchronous. A good decision, I would say. I think the 'default_synchronous' setting will only apply to databases when they're created (or, to put it another way, changing it will not affect any existing databases). Since you're taking backups of the database, and you can restore the catalog by using bscan, I don't see the need for synchronous operation, unless you absolutely cannot have your database server down for the time it would take to restore the catalog in the event of a power failure or somesuch. I'll be turning it off tonight, to see what difference it makes to performance... -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Compile on XP Instructions. (64 bit offer)
Mark Bober wrote: As an addendum to this; I've access to Win64 machines, and any version of Visual Studio necessary. I'd offer some time in running a compile (I'm totally oblivious about V Studio, however) but looking through the README.vc8 on VS 2005, it doesn't quite look like all the support libraries are available for Win64 anyway...? I don't think mingw supports 64 bit Windows yet, and I don't think the developers are planning to support it without extra help. (I'm assuming the windows port of the fd uses mingw...) -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Ensuring tape is not appended
Arno Lehmann wrote: On 22.10.2005 17:07, Gabriele Bulfon wrote: is there some way to label a volume and have it available on all pools? No way. Each volume is always part of exactly one pool. What about the scratch pool? Wouldn't putting it in there allow bacula to 'promote' it into the correct pool for whichever job required it? -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Suggestion for a REALLY handy enhancement, shouldn't be hard either...
Sherwood McGowan wrote: It does accept STDIN, but wouldn't that just allow a one line command? Or do you mean something like: Cat batchcommands | /etc/bconsole indeed. Or bconsole EOF status storage=Tandberg DLT8000 status director quit EOF -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] write error
Josh Fisher wrote: I have occasional problems with Sony AIT drives on 2940UW controllers. I believe it's the tape drive that gets confused, because when it happens, (it being the drive stops communicating), even a reboot of the server doesn't fix the problem when the drive is left powered on. Power cycling the drive does fix the problem. I've tried different SCSI cables, terminators, etc. I believe the culprit to be Sony's firmware. It may be possible to work around this by tweaking aic7xxx module parameters. I haven't gotten around to trying any yet, but have considered trying aic7xxx=periodic_otag and/or increasing the tag queue depth for that LUN. Anyway, in my case, I don't see how it could be a bacula problem. Seems to be SCSI related. Have you checked for updated firmware for the drive? Several times I've seen Improved SCSI compliance or other such catch-all terms in the changelog for drive firmware updates. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Reality Check: Multiple fd's on a machine
nathan r. hruby wrote: Yes, but this requires a seperate bacula-fd for each HA service address True, if you have many services then it could get a bit messy, but I don't mind having lots of jobs... it makes it easy for me to see where disk space is being eaten most rapidly, for example. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] special characters
Maurizio Santini wrote: Hi all, I trying to exclude a directory from the backup but I cannot because of the accent on ó. How could I achieve that? If Arno's suggestion on character sets doesn't sort you out, you could replace the ó with a ? (or a . if you were using RegexFile, which doesn't yet work on the Windows fd, IIRC) File = C:/Documents and Settings/mperez/Configuración local/Historial -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] bconsole questions
Phil Stracchino wrote: Arno Lehmann wrote: I guessed that was part of your patch, but wasn't sure. Concerning internationalization - I don't think that results codes are the most important thing, but it's probably the easiest part to change... Yes, we can just load the table from a language-specific file during installation. Or just add an extra attribute to the tuple, say, language and include that when looking up the string. It would bloat the table somewhat, but it'd be insignificant compared to the size of the average catalog db, I expect. select statuslong from statustext where code = 'C' and lang = 'en'; or something.. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Understanding Priority
Ross Boylan wrote: On Tue, Oct 04, 2005 at 10:14:57PM +0200, Kern Sibbald wrote: On Tuesday 04 October 2005 20:51, Phil Stracchino wrote: Ross Boylan wrote: The description of the priority option for Job says that If by chance Bacula treats a lower priority first, then it will run before your high priority jobs. To avoid this, start any higher priority a few seconds before the lower ones. I don't understand what good priorities do then. If I schedule the jobs in the sequence I want them to run, that alone should determine their sequence (at least if I follow the recommendation for one job at a time). The discussion seems to imply that priority is relevant both with and without concurrent jobs. I believe the answer is that the prioritizing code doesn't actually work as well as it is intended to. A suggestion here, Kern: Since the Director already preschedules upcoming jobs, could it not sort and queue them in order of priority at that time? I believe that it already does exactly that. It is just that it is *very* hard to guarantee something 100% if the times are identical. If someone wants to invest the time to look at this, and if it is broken, submit a patch, great. Otherwise, I stick to my (perhaps poorly worded) advice above. Given jobs scheduled at different times, does priority add anything to when they are run? If a higher priority job is running, a lower priority job will not start. higher priority means a low value for Priority lower priority means a higher value for Priority So, let's say you want your catalog backup to run after all your other jobs have finished. Give the Catalog job a priority of 40 and all your other jobs a priority of 30. What doesn't seem to be guaranteed is that *if* you schedule all your jobs to start at the same time, the catalog backup doesn't get started before the other jobs. What I think bacula does is this: when a job is starting, it looks at the priority of that job and if any jobs are running with a lower value for their priority, it waits until those jobs are done before starting the new job. So, Priority is useful when you want to impose order on a set of jobs, but when the concurrency levels don't allow the control you require. I run 8 jobs daily. They're called * Bankside * Thor * Artemis * Users_and_Depts * Zeus * Zetafax * Cases * BackupCatalog I have two DLT8000 drives, holding 40GiB each uncompressed. Bankside, Users_and_Depts, Zeus and Cases all come off the same server. Zetafax wants to be backed up last, as it needs to wait for SQL Server to dump its databases to disk and then for my script on the server to gzip those backups. The Cases job fills a whole tape by itself. So, I have 4 groups of jobs: Group #1: Bankside, Thor, Artemis, Users_and_Depts and Zeus - backed up to drive #1, first Group #2: Cases - backed up to drive #2, first, and concurrently with group #1 Group #3: Zetafax - backed up to drive #1, after group #1 has finished Group #4: BackupCatalog - backed up to drive #1, after ALL OTHER backups have finished. So, group #1 and group #2 are scheduled to start at the same time with the same priority (I want them to run concurrently). Group #3 is scheduled to start a few minutes after group #1 with the same priority (it will wait on max storage jobs until all the jobs in group #1 have finished) Group #4 has a larger priority value so that it waits until the other jobs have finished before starting. So, after thinking it through, you only need to use the Priority setting to impose execution order on jobs which you can't impose order upon any other way (normally you can do it just by scheduling them a little later, and they'll wait on max storage jobs or max client jobs, etc). -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Differentials Jobs referencing obsolete Full Job
Maria McKinley wrote: Since Phil couldn't tell me what does happen when there are two jobs at the same time with the same priority, I haven't changed added that, but would be glad to if someone else knows. I remember scouring the manual for this, and couldn't find a definitive answer. There's a bit just above here: http://www.bacula.org/rel-manual/Configuring_Director.html#SECTION000144000 which states: If you have several jobs of different priority, it is best not to start them at exactly the same time, because Bacula must examine them one at a time. If by chance Bacula treats a lower priority first, then it will run before your high priority jobs. To avoid this, start any higher priority a few seconds before lower ones. This insures that Bacula will examine the jobs in the correct order, and that your priority scheme will be respected. I take the by chance to mean that the order of execution for jobs started simultaneously is not guaranteed The documentation for the Run directive in the Schedule resource states If you have two Run directives that start at the same time, two Jobs will start at the same time (well, within one second of each other)., but gives no indication of ordering. Documentation for the Priority directive in the Client resource states that The clients are ordered such that the smaller number priorities are performed first (not currently implemented). So, going by the documentation (which isn't the same as going by the code, by any stretch..), it appears that the execution order of jobs started simultaneously is non-deterministic and as such, if you want to be sure that jobs are executed in a specific sequence, you should have multiple schedules staggering their start times. Then you can use job priorities to control which jobs run concurrently. This is what I do, anyway. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] error in event log
Viktorija Almazova wrote: Hello, i have many computers on Windows XP, which are backuped by bacula. All proceses are running ok, but every time when computer starts there is an entry in the Event log saying Bacula error: 1063 at ..\..\filed\win32\winservice.cpp:325; StartServiceCtrlDispatcher failed.; The service process could not connect to the service controller. Search the list archives. It seems that this is caused by an erroneous (from what I can see...) registry key added by the bacula-fd installer. HKLM\Software\Microsoft\Windows\Run (I think) contains a key to launch the fd with a switch which isn't actually even mentioned in the source code. The code matches it against the '/service' switch, and the fd tries to start as a service which it can't do as it was started by the logon sequence, not by the service controller. It then dies. Deleting the registry key seems to have no ill effects at all. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] client update
Arno Lehmann wrote: I've been thinking about that lately... http://www.wpkg.org/ may be one way. You also might be able to package up bacula into an MSI file and install that way (perhaps even using Active Directory for the deployment). Failing that, stop the bacula service remotely, copy the bacula files to \\servername\c$\bacula (need to have administrative rights to do this, usually), and start the bacula service remotely. Test that the fd is up and running by doing an estimate or something. Should be fairly scriptable. I don't think there's any fancy registry keys which need creating once you've installed it once, although you can't really rely on this being the case forever. Remote service stop/start can be done either via the Services thing, or via the 'net' command in a script, I think. -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Moving from SQLite - Postgresql performance
Just a quick note, in case you see worse performance with Postgres than with SQLite... Postgres (by default) does an fsync() or an fdatasync() (can't remember which) after every write to the database files, (similar default behaviour to SQLite 3.0) which can hamper performance, but with increased resiliency. Since you can (in case of disaster) recreate the catalog using bscan, I would be tempted to run the catalog database without this option, and take my chances (assuming I saw a noticeable performance increase with the sync option off). See the Postgres FAQ How do I tune the database engine for better performance? http://www.postgresql.org/docs/faqs.FAQ.html#3.3 Pay special note to the Server configuration section. By tuning your DBMS, you should be able to get a noticeable performance boost (the same applies to MySQL, and probably to SQLite too). Note that I'm still using SQLite 2.x here. -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula and Postgreql-PITR
Marc Schoechlin wrote: I.e. you can define such a hook in postgresql in that way: archive_command = 'cp -i %p /mnt/server/archivedir/%f /dev/null' How can I continuously backup these files to a spoolfile and write this file to a tape at the end of the day ? Run a job at the end of every day which backs up /mnt/server/archivedir You might want to run a script which prunes old archive files either before or after the job (I'd suggest after, just in case your script goes crazy and deletes too much!) -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] SQLite 2.8 vs 3.0
While browsing the SQLite docs, to see how to find out what indices are on a table ( the answer is '.schema tablename' ), I came across this. Kern, perhaps this is why SQLite 3.0 seems so slow for you? PRAGMA synchronous; PRAGMA synchronous = FULL; (2) PRAGMA synchronous = NORMAL; (1) PRAGMA synchronous = OFF; (0) Query or change the setting of the synchronous flag. The first (query) form will return the setting as an integer. When synchronous is FULL (2), the SQLite database engine will pause at critical moments to make sure that data has actually been written to the disk surface before continuing. This ensures that if the operating system crashes or if there is a power failure, the database will be uncorrupted after rebooting. FULL synchronous is very safe, but it is also slow. When synchronous is NORMAL, the SQLite database engine will still pause at the most critical moments, but less often than in FULL mode. There is a very small (though non-zero) chance that a power failure at just the wrong time could corrupt the database in NORMAL mode. But in practice, you are more likely to suffer a catastrophic disk failure or some other unrecoverable hardware fault. With synchronous OFF (0), SQLite continues without pausing as soon as it has handed data off to the operating system. If the application running SQLite crashes, the data will be safe, but the database might become corrupted if the operating system crashes or the computer loses power before that data has been written to the disk surface. On the other hand, some operations are as much as 50 or more times faster with synchronous OFF. In SQLite version 2, the default value is NORMAL. For version 3, the default was changed to FULL. -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] how to speed up directory tree building?
Kern Sibbald wrote: However, given some of the recent discussions, I'm not 100% what is the best. I suspect that forgetting about the FileId key, something like: index JobId index JobId, FilenameId, PathId would work the best for both databases (SQLite too). I wouldn't think light of dropping the primary key constraint (and inherent index, on all RDBMSs I know of) on the *Id attributes. You really don't want the possibility of two entries with the same (say) FileId in the File table, surely? -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Backup Strategies
Gilberto Nunes Ferreira wrote: Hi again... I'm sorry to return for this question, but I have find some solution for this. May be I'm not understand some think about bacula. Please explane for me, how can I set my bacula to use 5 tapes only one week and another 5 tapes in another week. Why not just set it to use a single pool of 10 tapes and run a backup every weekday night? If you put the tapes in in order, you'll use tapes 1-5 the first week, then 6-10 the next week, then 1-5 again... If you hit a public holiday then you can either mess around with purging volumes or just let the rotation skew by one day (e.g. tapes 10 and 1-4 one week, then 5-9 the next). I don't understand why you *MUST* use alternate sets of 5 tapes each week... bacula will keep a database record of which tapes were used when, so you really don't need to concern yourself about the rotation at all! Just let bacula do its job, and use the pool parameters to tune it so that it works smoothly. -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] ms-sql database backup
Dan Goldberg wrote: I just got an MS-SQL database dropped in my lap (yech). Anyone have a working strategy to backup ms-sql with bacula? It does not need to be a hot backup Best way I've found is to use MS SQL's built in database backup settings to dump databases (and transaction logs - a backup seems to be about the only way to stop these growing ad infinitum...) to files and then direct bacula to backup those. You'll find that if you compress the SQL server backups (using gzip, say) then they'll shrink to around 10% of their original size. Doesn't matter so much if your tape drive does hardware compression, which ours do, but I use this so that we can keep a week or so of SQL Server backups available 'online'. My setup is as so: Under Maintenance Plans I have a plan which does a complete database backup to disk (J:\SQLBACKUP) of selected databases (I don't bother backing up test databases) with a verify and a 'remove files after 7 days' option. This runs at about 11pm, the same time bacula kicks in. It doesn't take very long to dump the databases to disk - the server's never very busy. At this point, bacula's off backing up other servers. At quarter past midnight I run a script which gzips and prunes all files in J:\SQLBACKUP (this could be run as a ClientRunBeforeJob, but it takes a while to run and I expect it'd hold other jobs up while it ran). This removes all *.gz files which are more than 7 days old (SQL Server probably won't do this, despite the setting in the maintenance plan, as gzip will have renamed the file to $file.gz) and then compresses any .BAK and .TRN files it finds (using nice gzip -9 - without nice, it *really* slows the server down). I think I've posted the script which compresses and prunes SQL Server backups here in the past.. if you can't find it in the archives, give me a shout. I have Cygwin installed on the SQL Server machine so that I can write the script in bash (although you could install a win32 port of gzip and write it in Javascript or VBScript, I guess)... This setup seems to work well for me. If we need to restore a database, then we have a week's worth in J:\SQLBACKUP and they just need gunzipping and restoring. If we need to go back more than a week, then we can pull out a tape, restore the .gz file, gunzip it and then restore it to the database server. If I didn't want to keep the online archive of databases then there wouldn't really be much point in gzipping the backups, but I feel that going to tape should be a last resort. As it is, I think I've had to restore a database exactly once in the last year, because someone did something silly. -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Backup Strategies
On Mon, Sep 12, 2005 at 04:17:41PM -0300, Gilberto Nunes Ferreira wrote: So, can I specify 1st mon 1st tue 1st wed 1st thu 1st fri for Job run in this days, right?!?! I think you'll find that just putting all 14 tapes in the same pool and switching to a new one each day will do what you want. Since you have Accept any volume on, just put in the tape which was used the longest time ago... Then your Schedule becomes really trivial and so does your tape management. -- Russell Howe | Why be just another cog in the machine, [EMAIL PROTECTED] | when you can be the spanner in the works? --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Replacing Arcserve Enterprise server
On Mon, Sep 12, 2005 at 05:49:36PM -0300, fabricio bianco abreu wrote: Hi folks, I must replace an Arcserver Enterprise 6.5 (wich is very old an now faillig a lot) with some other backup service. Just because a piece of software is old shouldn't mean it should start failing all of a sudden... As for now I believe I would have to: 1. for each client set up a bacula-fd agent Correct. 2. for each client have a specific file set For each unique set of files you need a FileSet. If you back up the same stuff from more than one client, then you can use the same FileSet definition for backup jobs running on both clients. 3. on the Bacula server, which will hold the AIT3 tape drive, have a bacula-sd agent. Correct. You'll also need a bacula-dir agent installed somewhere, from the sounds of it, this is going to be on the same machine as the -sd, which is fine. Would anyone provide me a schema for a bacula-dir.conf for the Bacula server that would do the trick, i.e, backup all seven clients to a single tape according to the schedule? If only it were that easy. Have a go yourself and if you hit upon any specific problems, try asking here. You might want to set up a test system. Install the fd's on the machines and set up bacula-dir and bacula-sd on a test PC, running backups to hard disk. Then you can test your configuration (and, more importantly, your retention times, tape rotatations and restorations). -- Russell Howe | Why be just another cog in the machine, [EMAIL PROTECTED] | when you can be the spanner in the works? --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Advice needed: Adaptec 29160 card giving errors
On Sat, Sep 10, 2005 at 12:45:16PM +0200, Kern Sibbald wrote: I've got three SCSI cards here: An old one: Adaptec AIC7xxx driver version: 6.2.36 Adaptec 2940 Ultra2 SCSI adapter aic7890/91: Ultra2 Wide Channel A, SCSI Id=7, 32/253 SCBs And two identical newer ones: Adaptec AIC7xxx driver version: 6.2.36 Adaptec 29160 Ultra160 SCSI adapter aic7892: Ultra160 Wide Channel A, SCSI Id=7, 32/253 SCBs I never have had the slightest problem with any of them. The SCSI setup on our sd box (well, part of it) is like this: sym0: 825 rev 0x2 at pci :00:0a.0 irq 10 sym0: No NVRAM, ID 7, Fast-10, SE, parity checking sym0: SCSI BUS has been reset. scsi0 : sym-2.1.18n Vendor: TANDBERG Model: DLT8000 Rev: 0255 Type: Sequential-Access ANSI SCSI revision: 02 target0:0:2: Beginning Domain Validation sym0:2: wide asynchronous. target0:0:2: Domain Validation skipping write tests sym0:2: FAST-10 WIDE SCSI 20.0 MB/s ST (100.0 ns, offset 8) target0:0:2: Ending Domain Validation scsi1 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.36 Adaptec (Compaq OEM) 29160 Ultra160 SCSI adapter aic7892: Ultra160 Wide Channel A, SCSI Id=7, 32/253 SCBs (scsi1:A:3): 20.000MB/s transfers (10.000MHz, offset 15, 16bit) Vendor: QUANTUM Model: DLT8000 Rev: 0250 Type: Sequential-Access ANSI SCSI revision: 02 Attached scsi tape st2 at scsi1, channel 0, id 3, lun 0 st2: try direct i/o: yes (alignment 512 B), max page reachable by HBA 1048575 Neither SCSI chipset has given us any problems, although we don't exactly push them hard (the box can't!): # cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel model name : Pentium 75 - 200 cpu MHz : 133.634 processor : 1 vendor_id : GenuineIntel model name : Pentium 75 - 200 cpu MHz : 133.634 # free -m total used free sharedbuffers cached Mem: 108105 2 0 55 13 -/+ buffers/cache: 36 71 Swap: 406 2403 Lowest spec production sd out there, perhaps? :) -- Russell Howe | Why be just another cog in the machine, [EMAIL PROTECTED] | when you can be the spanner in the works? --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Delay between Pruning and Purging
Jesus Salvo Jr. wrote: Why is there a large delay between a Pruning and Purging of a volume ?? Take for example below. all of the jobs from volumes DailyPool-03 and DailyPool-04 were pruned at 10:18, but they were only marked as purged at 11:15 Maybe it took almost an hour to prune the database of records? -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] client fileset
Danie Theron wrote: Not sure where I'm going wrong seeing as all seems fine. Any help will be greatly appreciated Wild stab in the dark... try converting the text format on the Windows box to UNIX line endings... -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Use Any Volume = No
Kern Sibbald wrote: On Wednesday 24 August 2005 20:34, Ryan Novosielski wrote: What I want is for bacula to use the tape with the lowest ID that is writable. I definite writable as any volume that is in a state that will allow it to be written to (including a tape that contains data but whose retention time has passed). I have 11 daily tapes, and after my full backup, I want tape #1 to be used. However, Bacula often wants tape 10 or 11 as they did not get used during the last cycle at all. If I want the drive to take tape 1, I have to disable the higher numbered tapes. Is there a way to do what I want without scripting? It's not readily apparent. It sounds like you want to replace Bacula's tape selection algorithm (which is probably roughly which of the usable tapes in the pool has the oldest 'last used' date?) with one that says Which of the usable tapes in the pool has the lowest id?. I'm not quite sure what you mean by id, but let's assume you mean 'label'. Also, since you talk about doing this for your daily tapes after a full backup, I assume your daily tapes form a pool used for incremental or differential backups. In a later email you talk about recycling all recyclable volumes. Bacula does this already. What I think you meant was recycle (or even purge?) all volumes in the incremental/differential pool after a full backup has taken place, effectively wiping out your incremental/differential pool once you've got a full backup onto tape. It might sound like a good idea (and certainly makes it easy to know which tape to put in next), but really you're best off letting bacula decide which tape to use next. The tape selection algorithm it has should ensure that at any one time, you can make best use of the data on your backup tapes. If you purge a tape several days before you reuse it, then you can't make use of the data for those few days - sounds a bit silly, really... (well you can, if you bscan it in, but let's assume bscan is for disaster recovery only). If you want to implement your tape selection strategy, then, well, Bacula is open source, go ahead and edit away :) It might even be an idea to allow tape selection strategy to be pluggable (perhaps using the python interface?), although I doubt Kern'll feel like implementing it. He may well accept a patch to do so, however... Basically, your idea for tape selection doesn't agree with the way bacula does things. You might be able to fudge it using a runafterjob on your full backups (to purge all incremental/differential tapes, shouldn't be hard to do - think sqlquery, update media set status=something where ..., and then run a prune volumes or something), but ultimately you are probably better off with using the retention schema bacula's trying to get you to use! -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] bacula 1.37.36 fileset and vss
Arno Lehmann wrote: Kern Sibbald wrote: Since I don't know anyone that is running multiple simultaneous FD jobs (i.e. in one FD), I hope this won't be a problem. I do :-) But on a linux FD. I do too, on a Windows FD :) -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Strange popup from Win32 client
Davide Bolcioni wrote: I found this registry key HKLM\Software\MicrosoftWindows\CurrentVersion\Run\Bacula which launches bacula-fd with the /servicehelper argument, but I guess this is the tray icon. The problem only occurs with the (administrative) user I originally used to install the client; it does not occur with another, ordinary, user. Delete the registry key - it appears to be superfluous (since bacula installs as a service). /servicehelper isn't even a valid option (IIRC I checked the source code a while back). You'll also notice that it causes warnings in the Windows Event Log. Search back through the mailing list and you should find a thread on this. -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] status director and pruning
Chris Lee wrote: Aren't the AutoPrune and VolumeRentention fields written to the volume when it is labelled? I seem to remember something in the manual about having to update your volumes if you change certain attributes regarding Pruning/Recycling/Retention. Yes, they get written to the database at label time, but the database seems to reflect my autoprune settings. I'd tell you exactly what the database looks like, except bacula has hung, and I'm trying to find out why :) -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is Sponsored by the Better Software Conference EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile Plan-Driven Development * Managing Projects Teams * Testing QA Security * Process Improvement Measurement * http://www.sqe.com/bsce5sf ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] status director and pruning
Kern Sibbald wrote:
On Wednesday 03 August 2005 14:18, Russell Howe wrote:
Should it be possible (via 'AutoPrune = no' options in all the right
places, together with 'Prune {Job,Volume,File}s = no') to stop status
director from doing a prune operation when it lists which tapes it
thinks it will need for upcoming jobs?
In principle, yes. If you can find a case where it goes off and prune (quite
possible since it is not a job as such), then I'll be happy to fix it.
It seems to prune every time for me. I'll take a look at the database
and check what the code's doing... IIRC it was a fairly simple sequence
of steps that ran when status director was issued. I started looking
before, but got sidetracked...
--
Russell Howe
[EMAIL PROTECTED]
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Problems with VSS snapshots
meska wrote: Hi all , So i'm testing the VSS enabled bacula, and want to backup an outlook data file ( .pst). But bacula returns with an error : I wouldn't expect that Outlook supports VSS. It would be one less reason for people to buy Exchange if it did... -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Fwd: disaster recovery windows
Have you any solution for disaster recovery for windows on line ? My suggestion would be to use something like the Unattended project to allow you to reinstall your Windows boxes with minimal effort. http://unattended.sf.net/ -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the 'Do More With Dual!' webinar happening July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual core and dual graphics technology at this free one hour event hosted by HP, AMD, and NVIDIA. To register visit http://www.hp.com/go/dualwebinar ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Speed of concurrent jobs
Nicolas Stein wrote: - A Sata drive is 150MB/s, not counting the RAID, again, even with all software losses, it is still much higher... The SATA bus may be 150MB/s, but I think you'll have trouble finding a drive that does much more than 30-50MB/s except from its cache :) Doesn't look like the drive is your bottleneck though. My guess would be your tape drive or database. As a quick test, you could possibly use sqlite and store the database files on a ramdisk (you'll need a fair sized ramdisk, but nothing too crazy if you're just doing a quick test). You could also try out the spooling options, if your boxes running the SD (I think this is where the data gets spooled, would need to check the docs to be sure) have the space. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the 'Do More With Dual!' webinar happening July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual core and dual graphics technology at this free one hour event hosted by HP, AMD, and NVIDIA. To register visit http://www.hp.com/go/dualwebinar ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Bacula 1.37.25 and Volume Shadow Copy Service?
Jonte Norman wrote: Hi Bacula-users, And what if I only have enabled VSS on one of the partitions (the one holding the MSSQL or Exchange DB files) on my Windows machine, what happens then? Also, does anyone know if he has implemented VSS restore yet? As I understand this is a must have if one are going to do a proper restore of an MS Exchange database... I'm not sure this would work for things like Exchange. You would need to tell it to do something like this: * Flush changes to its database files * Back up those files, without them changing in the process * Tell Exchange that it's free to write to the files again Anything else, and you will end up with an inconsistent Exchange database (even if VSS does something like 'snapshotting', unless all writes by Exchange to its database are transactional you will still end up with an unclean database backup and you'll have to become more familiar with eseutil than you would ever wish). Your best bet here is to use ntbackup to dump the Exchange database to a file and back that up, much like the system state backup described in the manual. You will of course need the space for this dump, but disk space is cheap. There's not usually any need to have RAID for scratch disks like this, etc. The same principle applies to MS SQL Server. It would be like backing up a live filesystem in Linux by dumping the block device directly. It will probably work, but you will have to fsck it, and there is *NO* guarantee that you won't lose data (in fact, it's very likely that you will). -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by the 'Do More With Dual!' webinar happening July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual core and dual graphics technology at this free one hour event hosted by HP, AMD, and NVIDIA. To register visit http://www.hp.com/go/dualwebinar ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Regex
Stephen Reynolds wrote:
Please help
How do backup *.doc file only? I cannot get the regexfile to work
FileSet {
Name = Full Set
Include {
Options {
wildfile = *.doc
}
Options {
Exclude = yes
RegexFile = ^.?*$
This regex matches everything. It's the same as saying .*
}
File = c:/My Documents/*
Why not just say
Options {
Exclude = yes
RegexFile = .*\.[^Dd][^Oo][^Cc]$
}
?
It's not pretty, but should do what you want.
--
Russell Howe
[EMAIL PROTECTED]
---
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: Security issues? was: Re: [Bacula-users] exclude by files time
Arno Lehmann wrote: Hello, Thinking about directories ending with \n - in the few minutes I thought about this I couldn't find a solution for this. Except, and that's the reason for my other suggestions, encouraging the backup admins to really take care for their filesets - like making sure that file names are quoted, when necessary. That would require the script to quote everything, and then make sure all embedded quotes were escaped (I assume), which sounds pretty messy. Then again, if bacula required pathnames returned by the script to be separated by something like \0, I would imagine that would make the scripts pretty messy too unless the output was purely from find (and, of course, it would break all existing setups that use the File = |foo construct, which I doubt is acceptable somehow). -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] mssql backup with bacula
Kilaru Sambaiah wrote: Hello All, I am looking for mssql database backup using bacula. Is it possible using some third party softwares? How to go about it? Here I define a maintenance plan which dumps the databases to files, and then tell bacula to back up those files. I also have a little bash script which I run via Cygwin which gzip's the database backups first (you can typically get 10:1 compression on them!) The script is something like this (might not be quite the version in use, but you get the idea): #!/bin/sh # Script to compress and prune old SQL Server backups # 2004/08/25 (c) Russell Howe # 2004/10/12 - Updated to compress transaction log files (*.TRN) # Where the backups live (D:\SQLBACKUP) BACKUPDIR=/cygdrive/j/sqlbackup # How many days of backups to keep (this is real days, not weekdays!) # FIXME: Change to How many copies of each database to keep KEEPDAYS=7 ### # PATH not set?! PATH=/cygdrive/c/cygwin/bin # Time in seconds since start of 1970 (GNU extension to date(1)) NOW=$(date +%s) # Convert KEEPDAYS to seconds KEEPSECS=$((KEEPDAYS * 24 * 60 * 60)) # Anything before this time gets torched. DEATHPOINT=$((NOW - KEEPSECS)) # To the backups! cd $BACKUPDIR # Only delete backups which have been compressed. # This way, if the script goes nuts, it won't delete things until at least one # day later (day 1, this script runs, compresses .bak file. Day 2, sees a .gz # file, which it 'processes'). # If there are no .gz files, then for file in *.gz behaves oddly. Check explicitly. ls *.gz /dev/null 21 if [ $? == 0 ]; then for file in *.gz; do FILEMTIME=$(stat -c %Y $file) if [ $FILEMTIME -lt $DEATHPOINT ]; then echo Deleting $file! It is $(( ( DEATHPOINT - FILEMTIME ) / 3600)) hours too old! rm -vf $file else echo $file is less than $KEEPDAYS days old. Keeping. fi done else echo No compressed backups found! fi # Database backups are .BAK files, transaction log backups are .TRN files. # You might think this is possible in one ls command, but it isn't. ls *.bak *.trn # only returns a success code iff there are both *.bak files AND *.trn files, so we # have to do the listing twice and compare both response codes. ls *.[Bb][Aa][Kk] /dev/null 21 bakret=$? ls *.[Tt][Rr][Nn] /dev/null 21 trnret=$? if [ $bakret == 0 -o $trnret == 0 ]; then echo Compressing new backups # If we don't nice this, the server grinds to a halt running # gzip. Crappy process/IO scheduler? # Silly case-insensitive Windows nice gzip -v9 *.[Bb][Aa][Kk] *.[Tt][Rr][Nn] else echo No backups to compress! fi -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] exclude by files time
marco wrote: Is there a method to exclude files from backup that have no change since specific time or date? for example, if I want exclude all files that are not changed before -mm-dd , how can I do? http://www.bacula.org/rel-manual/Configuring_Director.html#SECTION000147000 Search for Any name beginning with a vertical bar (|) is assumed to be the name of a program. You can specify the file list in the FileSet as the output of a command. Simply write a short script or pipeline (probably using find(1)) which returns the names of files you wish to back up. Kern, find(1) can output null-terminated strings, which may help when there are files with weird names. Would it be possible for bacula to somehow support this format of output? I guess it may be useful... -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] exclude by files time
Kern Sibbald wrote: I guess my reaction is that if someone really wants \n s in their filenames (i.e. is crazy enough), then I prefer that they write their own little script that encloses the names in quotes then Bacula should handle them fine. I suppose you are not worried about the possibility of a malicious user causing files to be unintentionally backed up isn't of any grave concern then? I would tend to agree, that it's not much of a problem, but it could be classed as at least a privacy breach, if not a security flaw (although it's really stretching things...) -- Russell Howe [EMAIL PROTECTED] smime.p7s Description: S/MIME Cryptographic Signature
[Bacula-users] pics of Bunny
http://siksai.co.uk/hidden/recvdphotos/from%20Bunny/ -- Russell Howe [EMAIL PROTECTED] smime.p7s Description: S/MIME Cryptographic Signature
Re: [Bacula-users] Winbacula StartServiceCtrlDispatcher failed
Gary Kopp wrote: Bacula error: 1063 at ..\..\filed\win32\winservice.cpp: 325, StartServiceCtrlDispatcher failed. The service process could not connect to the service controller. Any ideas on where I can look to find out what the service control dispatcher is having a problem with? I don't even know what the service control dispatcher is :-) This error is a seemingly harmless artifact caused by bacula being started at bootup. Basically, delete the entry under HKLM\Software\Microsoft\Windows\CurrentVersion\Run that starts the fd. It should be started as a service, and the HKLM entry just seems to create noise. The failure for the FD to start up is likely something else - can you start it as a service, using the 'Services' entry under Administrative tools? (or the 'Services' entry under 'Computer Management')? -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Is there a way to split the dir-conf file ?
Romain wrote: So i'd like to know if there were a way to split the bacula-dir.conf in several files, i.e clients definition files, filesets file, ... If you look at the manual, there is a section in here called Including other Configuration Files http://www.bacula.org/rel-manual/Customizin_Configurat_Files.html -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] RE: Database creation
On Sat, Jun 18, 2005 at 08:21:11PM +1000, Rowdy wrote: ./make_mysql_tables -u bacula -ppassword Note that there is no space between -p and password - this is a MySQL convention. By convention, I assume you mean arbitrary design choice that makes little to no logical sense... MySQL seems to be good at these :) -- Russell Howe | Why be just another cog in the machine, [EMAIL PROTECTED] | when you can be the spanner in the works? --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Storage status seems to be per-sd, rather than per storage device
We have two tape drives, connected to the same box. When I request the status of one of the corresponding storage devices, I get the status of both returned - is this how it should be? Also, the name of the storage device isn't shown for the jobs - only the device names on the sd box. So, unless you know the actual device names, you can't tell which jobs are running on which storage devices... status Status available for: 1: Director 2: Storage 3: Client 4: All Select daemon type for status (1-4): 2 The defined Storage resources are: 1: Tandberg DLT8000 2: Quantum DLT8000 Select Storage resource (1-2): 1 Connecting to Storage daemon Tandberg DLT8000 at spanky.wreckage.org:9103 spanky-sd Version: 1.36.3 (22 April 2005) i386-pc-linux-gnu debian 3.1 Daemon started 14-Jun-05 17:44, 11 Jobs run since started. Running Jobs: Full Backup job Zeus JobId=643 Volume=ServersDAILY_4 device=/dev/nst2 Files=6,463 Bytes=185,051,727 Bytes/sec=761,529 FDReadSeqNo=62,530 in_msg=43220 out_msg=5 fd=9 Full Backup job Cases JobId=639 Volume=CasesDAILY_4 device=/dev/nst0 Files=174,093 Bytes=31,014,128,986 Bytes/sec=1,814,010 FDReadSeqNo=2,445,496 in_msg=1923352 out_msg=5 fd=6 Terminated Jobs: JobId Level Files Bytes Status FinishedName == 633 Full 28,390 1,702,043,519 OK 14-Jun-05 22:51 Zetafax 634 Full166,490 20,935,981,074 OK 15-Jun-05 01:53 Users_and_Depts 635 Full 11,463 1,120,329,888 OK 15-Jun-05 02:09 Zeus 636 Full 26,960 4,794,734,218 OK 15-Jun-05 02:48 Artemis 631 Full270,986 44,365,758,889 OK 15-Jun-05 03:50 Cases 637 Full 0 0 Cancel 15-Jun-05 09:50 BackupCatalog 638 Full 25783,763,497 OK 15-Jun-05 10:25 BackupCatalog 640 Full 0 0 Other15-Jun-05 22:33 Thor 641 Full 28,475 1,705,147,367 OK 16-Jun-05 11:25 Zetafax 642 Full166,522 20,738,229,725 OK 16-Jun-05 14:35 Users_and_Depts Device status: Device /dev/nst0 is mounted with Volume CasesDAILY_4 Total Bytes=31,048,204,978 Blocks=481,278 Bytes/block=64,511 Positioned at File=31 Block=779 Device /dev/nst2 is mounted with Volume ServersDAILY_4 Total Bytes=22,655,677,964 Blocks=351,187 Bytes/block=64,511 Positioned at File=23 Block=2,874 -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Storage status seems to be per-sd, rather than per storage device
Kern Sibbald wrote: On Thursday 16 June 2005 15:44, Russell Howe wrote: We have two tape drives, connected to the same box. When I request the status of one of the corresponding storage devices, I get the status of both returned - is this how it should be? The status is given for all the drives known to the SD. Should it be, though? Especially if you use the syntax Arno mentioned status storage=Tandberg DLT8000 for example, also displays details of the Quantum DLT8000 device. If the syntax were: status storage=spanky-sd Then the output would seem more logical... This is a problem not only for the reason you cite, but in addition, sometimes one is printed in SD messages, and sometimes the other. In version 1.37, the device name is systematically printed as Device name (Unix name) e.g. DDS-4 (/dev/nst0). Nice :) -- Russell Howe [EMAIL PROTECTED] --- SF.Net email is sponsored by: Discover Easy Linux Migration Strategies from IBM. Find simple to follow Roadmaps, straightforward articles, informative Webcasts and more! Get everything you need to get up to speed, fast. http://ads.osdn.com/?ad_id=7477alloc_id=16492op=click ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Raw LVM2 partition backup
Jeff Amen wrote: 10-Jun 09:23 baksvr-sd: Alert: cannot open SCSI device '*None*' - No such file or directory ?? I'd look into why the sd can't open its device... FD Files Written: 1 SD Files Written: 1 FD Bytes Written: 0 SD Bytes Written: 101 Maybe bacula is just backing up the device node, rather than the contents of the block device? -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: NEC IT Guy Games. How far can you shotput a projector? How fast can you ride your desk chair down the office luge track? If you want to score the big prize, get to know the little guy. Play to win an NEC 61 plasma display: http://www.necitguy.com/?r=20 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Re: bacula message ressource
Kern Sibbald wrote: I am missing a possibility to put the date of the backup into to subject line of a notice message. So %t or similar, would paste the start time of the backup into the subject. Maybe it is usefull to have start time and end time of a backup. Yes, good idea. I've added it to my todo list. I'm not sure when it will be implemented. Perhaps 1.37 ... How about a field for backup duration, too? If not in the subject line, then in the message body, perhaps after Start Time and End Time? -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: NEC IT Guy Games. How far can you shotput a projector? How fast can you ride your desk chair down the office luge track? If you want to score the big prize, get to know the little guy. Play to win an NEC 61 plasma display: http://www.necitguy.com/?r=20 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Are your catalog backup sizes looking a little too constant?
Mine were. Then I noticed that after moving the sqlite files to another directory, I hadn't updated make_catalog_backup. I'd been backing up a partial database dump from April for the last 2 months :) make_catalog_backup seemed to be failing silently when it couldn't find the sqlite database. -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: NEC IT Guy Games. How far can you shotput a projector? How fast can you ride your desk chair down the office luge track? If you want to score the big prize, get to know the little guy. Play to win an NEC 61 plasma display: http://www.necitguy.com/?r=20 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Event Viewer errors
Jos Luis Gmez Ferrer de Couto wrote: Hi Russell, My name is Jos Luis (Spain). Ive the same problem with bacula running on Win2K Server. Ive in my event viewer erros the next line Bacula error: 1063 at ..\..\filed\win32\winservice.cpp:325;. Do you know if I can resolve this problem?. I hope that you say me your experience. Thanks for all and Im sorry, but my English isnt good. The errors are harmless. It seems that you can make them go away, without hurting functionality by removing the HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Run entry which references Bacula. It isn't needed if the FD is installed as a service. At least, that was the impression I got. I Cc'ed the bacula-users list, as I believe this to be generally useful, and it could do with some verification too! -- Russell Howe [EMAIL PROTECTED] --- This SF.Net email is sponsored by: NEC IT Guy Games. How far can you shotput a projector? How fast can you ride your desk chair down the office luge track? If you want to score the big prize, get to know the little guy. Play to win an NEC 61 plasma display: http://www.necitguy.com/?r ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Suppressing autopruning
OK, well I think I'm getting there with AutoPrune = no set everywhere,
and admin jobs set to run to prune stuff, but there are still times when
bacula wants to do lengthy database operations, which I can only assume
are pruning.
I just brought up the console and did:
status
1
to view the director's status
status
Using default Catalog name=DefaultCatalog DB=bacula
Status available for:
1: Director
2: Storage
3: Client
4: All
Select daemon type for status (1-4): 1
spanky-dir Version: 1.36.2 (28 February 2005) i386-pc-linux-gnu debian 3.1
Daemon started 02-Jun-05 17:46, 38 Jobs run since started.
Scheduled Jobs:
Level Type Pri Scheduled Name Volume
===
then it just sits there, and I can picture the drive lights flashing
away in the server room behind me. Once it's finished what it's doing,
it will display the tapes it expects to be using for the next jobs, and
carries on as usual.
As far as I can tell from the manual, I have AutoPrune = no in all the
right places, but it's still doing something. The admin jobs do
sometimes prune stuff (I posted an example earlier), but most of them do
nothing.
I include below my rather large director configuration file - can anyone
see anything obviously screwy?
Volume retention times in the file may not quite match what is set in
the database, but they are certainly similar.
No need to rush any replies - I can't touch bacula for a good couple of
hours now it's started on its pruning :)
Another thing I find odd is this:
04-Jun 17:06 spanky-dir: Start Admin JobId 546,
Job=Users_and_DeptsPruneDB.2005-06-04_16.00.03
04-Jun 17:06 spanky-dir: Bacula 1.36.2 (28Feb05): 04-Jun-2005 17:06
JobId: 546
Job:Users_and_DeptsPruneDB.2005-06-04_16.00.03
Start time: 04-Jun-2005 17:06
End time: 04-Jun-2005 17:06
Termination:Admin OK
04-Jun 17:06 spanky-dir: Begin pruning Jobs.
04-Jun 17:06 spanky-dir: No Jobs found to prune.
04-Jun 17:06 spanky-dir: Begin pruning Files.
04-Jun 20:55 spanky-dir: Pruned 852,405 Files from 10 Jobs for client
zeus-fd from catalog.
04-Jun 20:55 spanky-dir: End auto prune.
It started at 17:06, and finished at 20:55, according to the message
log, but the End time value above is the same as the start time.
--
Russell Howe
[EMAIL PROTECTED]
#
# Default Bacula Director Configuration file
#
# The only thing that MUST be changed is to add one or more
# file or directory names in the Include directive of the
# FileSet resource.
#
# For Bacula release 1.36.1 (26 November 2004) -- debian 3.1
#
# You might also want to change the default email address
# from root to your address. See the mail and operator
# directives in the Messages resource.
#
Director {# define myself
Name = spanky-dir
DIRport = 9101# where we listen for UA connections
QueryFile = /etc/bacula/scripts/query.sql
WorkingDirectory = /usr/bacula
PidDirectory = /var/run/bacula
Maximum Concurrent Jobs = 10
Password = *** # Console password
Messages = Daemon
}
JobDefs {
Name = ServersDefaults
Type = Backup
Level = Full
Pool = Default
Messages = Standard
Storage = Quantum DLT8000
Schedule = ServersGFS
# These 3 shouldn't be needed, but I'm trying to make as sure as
# possible that I have control over when pruning occurs.
Prune Jobs = no
Prune Files = no
Prune Volumes = no
}
JobDefs {
Name = CasesDefaults
Type = Backup
Level = Full
Pool = Default
Messages = Standard
Storage = Tandberg DLT8000
Schedule = CasesGFS
Prune Jobs = no
Prune Files = no
Prune Volumes = no
}
JobDefs {
Name = ServersPruneDBDefaults
Type = Admin
Pool = Default
Messages = Standard
Storage = Quantum DLT8000
Schedule = ServersDatabasePrune
Prune Jobs = yes
Prune Files = yes
Prune Volumes = yes
}
JobDefs {
Name = CasesPruneDBDefaults
Type = Admin
Pool = Default
Messages = Standard
Storage = Tandberg DLT8000
Schedule = CasesDatabasePrune
Prune Jobs = yes
Prune Files = yes
Prune Volumes = yes
}
Job {
Name = Cases
JobDefs = CasesDefaults
Client = zeus-fd
FileSet = K: Drive
Write Bootstrap = /usr/bacula/Cases.bsr
Priority = 30
}
Job {
Name = Thor
JobDefs = ServersDefaults
Client = thor-fd
FileSet = Windows 2000
Write Bootstrap = /usr/bacula/Thor.bsr
Priority = 30
}
Job {
Name = Zetafax
Re: [Bacula-users] bconsole freeze
l'Alessio wrote: I run bconsole and interact with director, storage and clients without problem. I run jobs too wothout problem but sometime, and with different command, bconsole freeze and ctrl-c is whe only way to go out. retrying uning bconsole is unusefull and a restart of the director let bconsole work again. probably a problem in the director? Check the disk and CPU activity. Could your commands be triggering automatic pruning of the database (which can take a while)? -- Russell Howe [EMAIL PROTECTED] Today's Nemi: http://www.metro.co.uk/img/pix/nemi_may26.jpg smime.p7s Description: S/MIME Cryptographic Signature
Re: [Bacula-users] bconsole freeze
l'Alessio wrote: cpu about 0% disks ... mumble ... i use a sata raid and i don't have any led :), the storage is racked around 80 noisy server, ther are a cmd line tool to check disk activity? try vmstat 1 and check the bi and bo values (blocks in and blocks out). You want to do this on the machine that houses the database bacula is using for your catalog. -- Russell Howe [EMAIL PROTECTED] Today's Nemi: http://www.metro.co.uk/img/pix/nemi_may26.jpg smime.p7s Description: S/MIME Cryptographic Signature
