On Tuesday, 04 February 2020 17:58:26 SAST Martin Simmons wrote:
> It looks like your Storage Daemon (TapeServer) is runnning an older Bacula
> version than your Director (trooper-dir). Bacula expects the versions to be
> the same.
Thanks, all working fine after yum update.
Cheers, Ian
--
On Tuesday, 04 February 2020 17:58:26 SAST Martin Simmons wrote:
> It looks like your Storage Daemon (TapeServer) is runnning an older Bacula
> version than your Director (trooper-dir). Bacula expects the versions to be
> the same.
Ah. I thought it was perhaps some piece of hardware like a
hi all
Can someone who knows please advise if this is an issue with my database, my
tape server, the tape machine, or the tape?
The director machine (this machine) died (power failure) without being shut
down properly.
Tape server is HP, as is tape drive. Tape was out of drive, and I shut
On Saturday, 01 February 2020 00:15:07 SAST you wrote:
> I've tried a few times, including restarting MariaDB and the tape machine.
I mean Postgresql. Keep forgetting Bacula runs with Postgresql :-)
Thanks, Ian
--
i...@zti.co.za http://www.zti.co.za
Zero 2 Infinity - The net.works
Phone
On Monday, 03 June 2019 11:44:34 SAST Tilman Schmidt wrote:
> It's by design. The "Error" state for a tape means it's defective and should
> not be used anymore. If you want Bacula to use it again (and i you are sure
> what you are doing) you have to change its state manually from "Error" to
>
hi all
I suspect this has been asked before but can't find the answer ... maybe using
the wrong words on Google.
See attached, I was expecting Bacula to reuse this tape this morning, but
instead it demanded a new one.
Why would that be?
I tried to find the procedure for dealing with "error"
hi
My other message re error tapes refers, here's the bconsole session for
labelling the new one, as far as I know it was a brand new tape, so am
wondering if the error message below is a program bug.
It first says it is unlabelled and then says it was pre-labelled. ??
On Wednesday, 23 May 2018 3:58:24 PM SAST Ian Douglas wrote:
> Let me try and upgrade Bacula on the SD box and see if that fixes things.
> Hope the CentOS version has been updated.
Yum update and restarted SD, just in case.
A happy ending to test backup of catalogue.
Termi
hi
On Wednesday, 23 May 2018 3:44:32 PM SAST you wrote:
> bacula-dir and bacula-sd always must be in the same version
I suspected as much (from previous battles) but I checked status via Bat and
director seems to talk to SD okay so I thought it was maybe backwards-
compatible.
Let me try and
hi
Google won't tell me the answer so I need to ask the experts...
Looks like my Bacula on this box (director, fd, and database via Postgres)
updated.
Now I get errors like this:
23-May 14:09 trooper-dir JobId 18205: Start Backup JobId 18205,
Job=BackupCatalog.2018-05-23_14.09.13_08
23-May
On Sunday, 18 March 2018 5:41:14 PM SAST Phil Stracchino wrote:
> There's two issues here:
> 1. MySQL has historically allowed columns to be declared NOT NULL
> without a DEFAULT. This is wrong and has always been wrong (because
> you're telling MySQL a column may not be left empty, then not
On Monday, 12 March 2018 11:57:49 PM SAST Josip Deanovic wrote:
> Maybe in .local/share/local-mail
The trash is stored there now (as opposed to the previous Trash folder).
Still no sign of where the rest is ... asked on Gentoo support as well as on
KDE bug list ...
Thanks, Ian
--
On Monday, 12 March 2018 11:57:49 PM SAST Josip Deanovic wrote:
> Maybe in .local/share/local-mail
There is a mail structure there but last written 27 Jan, same as other folder.
I think Kmail messed around in that folder when it upgraded.
It's a real mess ... the messages are in the same
hi All
On Monday, 12 March 2018 7:17:15 PM SAST Martin Simmons wrote:
> Also check that some files under
> /home/ian/.local/share/.local-mail.directory have changed (possibly mail is
> being written somewhere else now?).
Ah, I overlooked the obvious. Yes indeed it does appear as if Kmail in its
On Tuesday, 09 January 2018 9:43:02 PM SAST Jeffrey R. Lang wrote:
> What version of bacula are you running? I had a lot of issues with
> version 7.4 with these types of errors. Once I upgrade to 9.X all these
> error went away And I have had not further SD failures. If you're your
> not
On Tuesday, 09 January 2018 9:02:16 PM SAST Ian Douglas wrote:
> The Inc job uses different tapes to Full job, but Director is not requesting
> tape change because it's waiting for SD to say its finished, I suppose...
Looks like it eventually timed out and has requested the tape change.
hi All
This is annoying: literally half way through a 3.6 TB backup.
09-Jan 19:20 TapeServer JobId 15020: Writing spooled data to Volume.
Despooling 50,000,033,879 bytes ...
09-Jan 19:32 TapeServer JobId 15020: Despooling elapsed time = 00:11:35,
Transfer rate = 71.94 M Bytes/second
09-Jan
t's waiting for SD to say its finished, I suppose...
Thanks, Ian
On Tuesday, 09 January 2018 8:26:08 PM SAST Ian Douglas wrote:
> hi All
>
> This is annoying: literally half way through a 3.6 TB backup.
>
> 09-Jan 19:20 TapeServer JobId 15020: Writing spooled data to Volume.
Hi all
I updated my version of Postgres. Foolishly did it while Bacula was busy with
a long backup, but that's another story.
I'm on Gentoo Linux and followed their upgrade guide.
Anyway, normal backups run fine. However the nightly catalogue backup fails
thusly:
07-Nov 23:10 trooper-dir
On Thursday, 08 June 2017 2:49:52 AM SAST Ivan Adzhubey wrote:
> Looks like your directory is set up to automount and Bacula does not know
> how to handle it properly. You might need to write a small Bash script
> doing just what you tried manually: "cd /home/ian/nas1stuff/; ls -lh" and
> then
On Saturday 24 June 2017 14:19:13 Jari Fredriksson wrote:
> I'm a Linux veteran but now having one FreeBSD instance too.
>
> What is the proper way to install an bacula-fd into this. # pkg install
> bacula-fd or bacula-client did not work.
FWIW when I was in a similar situation trying to get
Hi all
Lately I've been getting errors like
Could not stat "/mnt/nas1stuff/": ERR=Host is down
where I'm trying to back up files from a NAS box (FreeNAS), and the target
folder is mounted on this box, as was recommended to me on this list.
So the folder does not get backed up.
If I run the
On Thursday 05 January 2017 23:25:59 Ian Douglas wrote:
> The messages do not say WHICH server, I presume Director? But it is
> running
I restarted Director, then reran the job, it picked up a 2 file discrepancy
between database and tape which it corrected, and now seems to be r
On Wednesday 04 January 2017 07:15:42 Josh Fisher wrote:
> > Now the job that WAS running (and all ones after it) are Error.
>
> Yes. So you cannot trust the job that was running. I would probably
> restart all daemons and erase that tape and start over.
Thanks. Was hoping for a more elegant
Hi All
Compliments of the season and all that.
Year is off to a rocky start. :-)
Was doing a full backup, around 3.5 to 4 TB so takes a while.
PC running Director decided to lock up for some reason (possibly related to
new CPU fan).
Now the job that WAS running (and all ones after it) are
On Wednesday 26 October 2016 10:20:33 Simone Caronni wrote:
> Please try the repository contains a rebuild of the official packages for
> RHEL/CentOS and Fedora.
>
> https://copr.fedorainfracloud.org/coprs/slaanesh/Bacula/
>
> Upgrade is straightforward from CentOS 6/7 packages and does not
On Tuesday 25 October 2016 21:35:25 Ian Douglas wrote:
> yum update on Centos did not install new update. I tried the remi repo.
> Already have epel.
I see I also have this repo:
epel-bacula7/7/
hi All
On Thursday 07 July 2016 15:03:00 Kern Sibbald wrote:
> I am pleased to announce that we have just released Bacula version 7.4.2.
My Gentoo box running bacular-dir updated to 7.4.3, however the storage
director on CentOs 7 is sitting at 7.0.4 and dir won't talk to sd ...
yum update on
On Sunday 02 October 2016 18:56:31 Kern Sibbald wrote:
> For simple "manual" things such as what you are doing, it is *far*
> easier to use bconsole. Bat is far more useful when you are looking at
> a number of Volumes or Job reports, ... Bat is not particularly good
> at doing console type
On Sunday 02 October 2016 14:23:18 Ian Douglas wrote:
> I've gone around in circles with mount, unmount, label, including
> physically removing the tape after unmount, but not getting anywhere.
Curiously, it has never asked me to specify the pool.
I've tried restarting bat but c
hi All
Can't believe I'm having this runaround but I am.
Tape drive needed a new tape. So I unmount, put in new tape, and mount.
Guess I should have 'label' first.
But anyway, now Bacula sees the tape but insists it is unlabelled.
Automatically selected Catalog: MyCatalog
Automatically
hi Ana
On Tuesday 19 July 2016 15:37:05 Ana EmÃlia M. Arruda wrote:
> If you run a differential after a full, then an incremental, but the
> differential one hadn't finished before the incremental one starts, then
> the incremental would check the last full one. So you will have both
>
Hi All
It seems to me that Bacula makes unnecessary duplicate backups.
I have
1. Daily incremental
2. Monthly differential
3. Annual full.
However one of the backups is over 3.5 TB, and the system only manages about
100GB/hour.
So the differential backup (in this case) ran for more than 24
hi All
On Thursday 21 April 2016 17:00:14 Clark, Patti wrote:
> If you are using spooling (recommended) your spool size parameters are used
> to control the size of your job spool and the total spool space available.
> The example below shows 50GB of job spool size and a total of 1TB of spool
>
hi All
On Thursday 21 April 2016 07:21:08 Josh Fisher wrote:
> Yes, most likely. In the 3 hours or so it took to despool to tape, the
> SD->FD TCP connection was dropped for some reason. I am of the opinion
> that not all interfaces interpret IEE 802.3az Energy-Efficient Ethernet
> in the
hi All
Is there a way to recover from this situation?
I'm trying to make up a NAS drive with many TB of data. 4th attempt.
I don't know why it lost connection to the SD, box is up and running and has
not gone down:
ssh tapeserver
ian@tapeserver's password:
Last login: Thu Apr 21 08:18:40
hi All
Kern thinks it is not a bug.
Can someone please advise what Bacula does when one or more of the machines
(eg SD, FD or Director) goes down while a job is running?
thanks, Ian
On Monday 04 April 2016 14:37:15 Ian Douglas wrote:
> hi
>
> I'm a new user ... hi to all.
>
hi
I'm a new user ... hi to all.
Was busy with my initial backups, had some issues, and now Storage wants to
continue a backup on a non-labelled tape:
Writing: Full Backup job Backup2Nas-to-Tape JobId=63 Volume=""
How do I tell it to use the volume that it originally wanted, and that is now
38 matches
Mail list logo
