--On Friday, December 04, 2015 08:19:00 AM -0600 Richard Robbins
wrote:
> The OS mounts the NFS share at that point and I'm able to read and write
> files without difficulty but when I fire up Bacula the program hangs with
> accompanying warning messages "Warning:
--On Thursday, November 19, 2015 01:03:59 PM + Martin Simmons
wrote:
> Does Bacula ever check for expired [data encryption] certs? I suspect
> not, so the question about rollover strategy is a moot one.
I've empirically verified this to be the case; I performed a
--On Thursday, November 19, 2015 10:49:07 AM +0100 Marcin Haba
wrote:
> You can renew your certs.
True, as long as you're ok with using the old key. However t won't work,
for example, if you need to expand your key size.
> I think that important is understand that data
My alerting system tells me that I have some file daemons that have been
merrily encrypting their data for quite a while. In particular, the
expiry dates for the data encryption x509 certs are coming up soon.
Well, this brings up an interesting question that I'd not really
considered in depth:
--On Thursday, May 21, 2015 12:21:23 PM -0500 Dimitri Maziuk
dmaz...@bmrb.wisc.edu wrote:
How much work is required to mount a filesystem on such drive? That is,
one drive from md raid1 with extX on it just mounts when you stick it
in a usb cradle. Do you need to jump through extra hoops with
--On Thursday, May 21, 2015 06:50:31 PM +0200 Rados?aw Korzeniewski
rados...@korzeniewski.net wrote:
Why do you need to use a 500MB volume in size? This days it is like
distributing movies on floppies instead of DVD/BR.
Some of my older deployments had data patterns where incrementals
are
--On Thursday, May 21, 2015 09:06:41 AM +0200 Kern Sibbald
k...@sibbald.com wrote:
Bacula does keep 64 bit addresses.
Excellent. Not surprisingly, I'm not dealing with file sizes near 2^63,
but I *do* need to back up files that are in the 2^39 range (from
filesystems that are in the 2^46
--On Thursday, May 21, 2015 11:09:58 AM -0500 dweimer dwei...@dweimer.net
wrote:
I have three systems two of which are using disk backup, then copy to
tape both of those are running on CentOS with 25G volume sizes [...]
The third system is using 46G File volumes
Thanks. Good to know.
I
I was under the impression that the maximum size of a file
that can be backed up would be either 2^63 or 2^64 bytes, but
I can't seem to find anything in the manuals or via google-fu
that confirms this.
Does anyone have any positive information regarding the
maximum file size limit?
Devin
--On Monday, January 26, 2015 01:46:07 PM -0600 ramesh penugonda
ramesh.penugo...@gmail.com wrote:
I am getting the encryption context error on a freebsd server when trying
to backup data encrypted.
1) when i run in daemon mode i get the encryption error
26-Jan 19:33 bksrv1-dir JobId
Also check both your Solaris and attached switch statistics
for the *late* collision count. If it is anything but zero,
it is indicative of an autonegotiation problem. Solaris
100Mb interfaces are infamous for needing autoneg forced off
and locked at a set value at both the server and switch
I have a situation where I have one bacula installation locally
for normal backups and another at a geographically remote
location for DR (disaster recovery) purposes.
For the DR site I would like to minimize network traffic, so I've
set things up such that I manually triggered one Full backup of
--On Thursday, January 20, 2011 09:27:37 PM -0500 Dan Langille
d...@langille.org wrote:
On 1/17/2011 4:27 PM, Devin Reade wrote:
Eh? Have I missed some important point about using MySQL for Bacula's
catalog, or is this just the author's prejudices speaking?
As the author of the above
--On Monday, January 17, 2011 10:50:39 AM -0500 Dan Langille
d...@langille.org wrote:
(In another thread)
Even though you are doing migrate, this might help, because Migrate and
Copy are so similar.
http://www.freebsddiary.org/bacula-disk-to-tape.php
I started to scan that document,
Maybe try looking at your database and seeing if there are any
locks still in place?
--
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their
I have a situation where I have two geographically separate
networks, both of which are using bacula locally. I'll call
them network A and network B. Currently, (bacula) server A
on network A backs up network A machines, and server B on
network B backs up network B machines.
I'd like to
--On Friday, January 07, 2011 12:23:43 AM -0700 Devin Reade g...@gno.org
wrote:
I'd like to (additionally) set things up so that server A
performs offsite backup for network B, and server B does the
converse for network A (for DR purposes).
I'm also open to other configurations, such as having
Martin Simmons mar...@lispworks.com wrote:
I read that glusterfs uses FUSE, so it might be checking something more than
the uid. That would explain why a root shell can access the files. Note that
the error is Operation not permitted, which is different from the normal
Permission denied
I've got a Linux-HA resource agent for bacula-fd that is suitable
for backing up clustered filesystems. I've cc'd the pacemaker list
and mentioned that anyone is welcome to incorporate it into the
resource-agents RPM.
The RA is available from:
ftp://ftp.gno.org/pub/tools/bacula-contrib
Martin Simmons mar...@lispworks.com wrote:
On Wed, 22 Dec 2010 14:09:50 -0700, Devin Reade said:
I have set up bacula clients on these nodes and, in addition to the
usual ext3 filesystems (/, /usr, et cetera), I'm trying to back up
the glusterfs-mounted /home, however I'm seeing the current
That's kernel space, not application space. Your most likely culprits
are bad memory or excess heat. Try to log/graph lmsensors for the
latter, and run memtest86 (for at least one full run) for the former.
Also look at any other hardware monitors you might have available,
including smartd.
Yes, this is bacula related, but first some background.
I've got a new two-node HA cluster where I am trying the new (for me)
mechanism of using glusterfs for /home. (For anyone not familiar with
this, both nodes have native filesystems mounted elsewhere -- in this
case, /gluster/home -- and
Thank you very much for doing the work of making this release (in
particular, the el5 release which I am using on CentOS 5).
It doesn't appear that the bacula-bat RPMs are up on sourceforge in
the fschwarz directory as they were with previous releases. I'm assuming
that this is related to the
Over the years I've learned that the ability to read DVDs written
on one a different drive than the original writer is much improved
if the disk is written at speed=1. I recently set up a DVD on my
SD and would like to continue this level of paranoia.
I was looking through the dvd-handler script
Piotr Gbyliczek p...@forlinux.co.uk wrote:
On Sunday 14 December 2008 04:52:50 Devin Reade wrote:
If you have clients that are not on the same network as one of the
SD interfaces (or if you have clients with dumb resolvers that won't
do result record sorting based on connected interfaces
Piotr Gbyliczek p...@forlinux.co.uk wrote:
I'm having problem with configuring bacula properly. We have one director,
three storages and lot of clients. Clients are in different networks, so we
need to use different IP to connect from client to storage and different IP
to connect from
Magnus Ahl [EMAIL PROTECTED] wrote:
Our server came with blank tapes so there were no problem with labeling. How
about unmounting them from bacula, feed them to the drive manualy using mtx
and erasing them with /bin/mt -f /dev/nst0 erase?
Assuming that it is in fact okay to destroy the
I'm just in the process of setting up bacula and debating postgres
vs mysql (I've used both, but have more experience with the latter).
I see that the current online users' manual mentions that one advantage
of using Postgres is support for transactions and stored procedures.
MySQL also has
Arno Lehmann [EMAIL PROTECTED] wrote:
Well, I suppose lack of transactions can make MySQL sound less reliable,
but unless your catalog server tends to crash or damage its filesystem I
don't think this is the most important issue...
I was thinking more of the situation where concurrent db
29 matches
Mail list logo