Re: [Bacula-users] [MTX Tape Changer] Error While Unloading Tape from Drive

I can imagine several possible problems: 1. You may be using the tape drive rather than scsi the control channel. 2. VTLs do not always emulate hardware correctly. 3. You may be using a scsi control channel that points to something other than an autochanger. Best regards, Kern On 02/18/2016

Re: [Bacula-users] Volumes are not being recycled?

Hello Ana, It may be possible that your code would work. However, please be aware that it was *very* complicated to get the current code to produce the right results. It is very easy to get an answer that is wrong by 1 hour or something like that. Thus anything that one would submit would ne

Re: [Bacula-users] Volumes are not being recycled?

Hello, Yes, it is the list_result() function in src/cats/sql.c that is a very generic routine for outputting "list" results. I personally cannot be of much help, but there are a good number of SQL experts using Bacula. Best regards, Kern On 02/18/2016 10:29 AM, Wanderlei Huttel wrote: > Hello

Re: [Bacula-users] Volumes are not being recycled?

Hello, That may be a solution. I am not particularly worried about SQLite as it is no longer supported. However, what I do need is a patch that will work for both MySQL (MariaDB as well) and PostgreSQL. Some sort of stored function may also be appropriate. Best regards, Kern On 02/18/2016 10

Re: [Bacula-users] onefs query

Hello, I have some additional knowledge. I thought maybe the problem was that onefs should be in the "default" options block. But if I change the fileset to this, the client segfaults when running the estimate command on it: FileSet { Name = "Machine Heavy-Weight NO RDBMS" Include {

[Bacula-users] [MTX Tape Changer] Error While Unloading Tape from Drive

Hi All, Good day, I have a problem using MTX Tape Changer, I failed to unload the tape from drive, I'm using VTL (data domain) as library, MTX as tape changer, and solaris as the OS, This is the error message when I try to unload the tape : -bash-3.2$ /usr/local/sbin/mtx -f /dev/scsi/changer/c4t

[Bacula-users] onefs query

Hello, I'm using bacula 7.2.0 and I have a fileset that looks like this: FileSet { Name = "Machine Heavy-Weight NO RDBMS" Include { Options { # DANGER: This will allow this FileSet to cross file system # boundaries. onefs = no fstyp

Re: [Bacula-users] Volumes are not being recycled?

Hello Kern, Thank you for your clear explanation. I was thinking about an "ExpiresAt" field instead of an "ExpiresIn". The meaning of an ExpiresIn field is perfectly clear for me expressed in seconds. Also, I agree that the results from bconsole commands should be less human-readable since we have

Re: [Bacula-users] Volumes are not being recycled?

Hello Guys I was trying to create a function to convert seconds and bytes to more "human readable". If I'm right, the file that shows information in bconsole is src/cats/sql.c, function list_result. How the function list_result didn't check the name of fields, I guess is necessary to check the na

Re: [Bacula-users] Volumes are not being recycled?

On 02/17/2016 05:58 PM, Kern Sibbald wrote: > Hello Heitor, > > One of the basic design conditions for Bacula is that as much as > possible that goes into it officially should be understandable and > maintainable by an average C/C++ programmer. Once things start getting > complicated, I slow d

Re: [Bacula-users] Volumes are not being recycled?

Hello Heitor, One of the basic design conditions for Bacula is that as much as possible that goes into it officially should be understandable and maintainable by an average C/C++ programmer. Once things start getting complicated, I slow down ... Best regards, Kern On 02/18/2016 08:25 AM, Hei

Re: [Bacula-users] Faster reading of Volumes from disk

> On Feb 17, 2016, at 9:57 AM, Martin Simmons > wrote: > >> On Tue, 16 Feb 2016 18:59:10 -0500, Dan Langille said: >> >> As a test, I scp'd over all the volumes for one job and timed it: >> >> real167m15.956s >> user117m51.006s >> sys 23m56.653s >>

Re: [Bacula-users] Volumes are not being recycled?

Kern, Sorry for the dumb question but could not bconsole make the values conversion in similar way to this shell script bytes conversion snippet before printing? xargs -i echo 'scale=2; {}/1073741824' | bc May it would be a more database independent approach. Regards, =

Re: [Bacula-users] Faster reading of Volumes from disk

scp can move binary copies of blocks very efficiently. A copy (or migrate) job reads one record (generally smaller than blocks) one at a time, and repacks them into a new block. The block is then checksummed (takes significant time) before it is written. Look at your CPU usage that should la

Re: [Bacula-users] Volumes are not being recycled?

Hello Ana (and Heitor), Please note that at the current time, the list command is generic, and thus it only knows how to print character strings that have been returned by the SQL engine. Though there may be some way to tell SQL that the result we get for expiresin is a "duration" (note: very

Re: [Bacula-users] Faster reading of Volumes from disk

> On Feb 17, 2016, at 1:49 AM, Andrey Tataranovich > wrote: > > On Tue, 16 Feb 2016 15:21:27 -0500 > Dan Langille wrote: > >> I can scp a 5GB Volume from one system to another in about 90 seconds. >> >> Why does that take the SD 8-10 minutes to do the same? >> >> The two systems are on the s

Re: [Bacula-users] Faster reading of Volumes from disk

> On Feb 17, 2016, at 7:57 AM, Josh Fisher wrote: > > > On 2/16/2016 6:59 PM, Dan Langille wrote: >>> On Feb 16, 2016, at 3:21 PM, Dan Langille < >>> d...@langille.org > >>> wrote: >>> >>> I can scp a 5GB Volume from one system to another in

Re: [Bacula-users] Faster reading of Volumes from disk

> On Tue, 16 Feb 2016 18:59:10 -0500, Dan Langille said: > > As a test, I scp'd over all the volumes for one job and timed it: > > real167m15.956s > user117m51.006s > sys 23m56.653s > > FWIW, these copies were done while a ZFS scrub was underway, so if anything, > the potential

Re: [Bacula-users] Faster reading of Volumes from disk

On 2/16/2016 6:59 PM, Dan Langille wrote: On Feb 16, 2016, at 3:21 PM, Dan Langille > wrote: I can scp a 5GB Volume from one system to another in about 90 seconds. Why does that take the SD 8-10 minutes to do the same? The two systems are on the same 1Gb/s network.