Re: How testing a new backup system in parrallel with the old one ?
On Wed, Sep 14, 2005 at 02:35:27PM +0200, rangzen wrote: > Hello, > > after a week of test, everything seems fine. > > So it seems that's the good way : "That is the key. In your disklist > file you can name your entries distinct from the path." > > Thank you very much ! In your new config you may have set "record no" for testing. When you switch over, don't forget to match it to your other config. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: How testing a new backup system in parrallel with the old one ?
Hello, after a week of test, everything seems fine. So it seems that's the good way : "That is the key. In your disklist file you can name your entries distinct from the path." Thank you very much !
Re: Can someone explain to me what no-hold in amstatus means ?
On Wed, Sep 14, 2005 at 12:50:20PM -0400, Guy Dallaire wrote: > 2005/9/14, Jon LaBadie <[EMAIL PROTECTED]>: > > > > Just the dump space. Backup the config dir and other associated dirs. > > Some would actually reflect the previous run as they are not complete > > until the end of the dump. > > > > > If I lose the index data, it means I will no longer be able to use > > > amrecover right ? > > > > Correct, though amrestore or manual recovery will still work. > > > > I've just realized that maybe my understanding of the holding disk > concept was not adequate. It looks like IDEALLY, the holding disk(s) > devices should be used/dedicated specifically for that purpose only. Two purposes I know of for the holding disk: - a place to collect your dumps when a tape drive is not available, either intentionally (let them collect and then flush) or from some error state - enable multiple DLE dumps to progress at the same time, parallelism. they collect on the holding disk until complete while taper picks one completed DLE dump at a time to send to the tape > In my case, the holding disk was used as the holding disk, as well as > a place for amanda db, logs and indexes. The disk was probably VERY > HEAVILY STAMPED ON so that might explain the no-hold problem. It is definitely not intended for config/log/index/debug file data. But, you had said you had 130GB of free holding disk space available and only 8GB of dumps to do. Even with the config ... data you should not have run out of space. > P.S.: I'll leave the holding disk device OUT of the disklists as I > don't think it need to be backed up if it only contains the amanda > dumps. Right ? I never have. Do you want a dump of your dumps? :) -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Can someone explain to me what no-hold in amstatus means ?
2005/9/14, Jon LaBadie <[EMAIL PROTECTED]>: > > Just the dump space. Backup the config dir and other associated dirs. > Some would actually reflect the previous run as they are not complete > until the end of the dump. > > > If I lose the index data, it means I will no longer be able to use > > amrecover right ? > > Correct, though amrestore or manual recovery will still work. > I've just realized that maybe my understanding of the holding disk concept was not adequate. It looks like IDEALLY, the holding disk(s) devices should be used/dedicated specifically for that purpose only. In my case, the holding disk was used as the holding disk, as well as a place for amanda db, logs and indexes. The disk was probably VERY HEAVILY STAMPED ON so that might explain the no-hold problem. What I did, is that I moved the idx, log and db directories to /usr/local/var/amanda (as explained in the sample config file section that I probably failed to read correctly or skipped over) and left ONLY the holding disk directory on the physical disk. This should speed up things considerably. Again, thanks again for your time P.S.: I'll leave the holding disk device OUT of the disklists as I don't think it need to be backed up if it only contains the amanda dumps. Right ?
Re: Can someone explain to me what no-hold in amstatus means ?
On Wed, Sep 14, 2005 at 11:04:59AM -0400, Guy Dallaire wrote: > 2005/9/14, Jon LaBadie <[EMAIL PROTECTED]>: > > > > An alternative, if you use gnutar rather than dump, is to exclude > > the directory containing the holding disks space. I have my > > holding disk spread over 4 drives. I named each consistantly > > as dumps/amanda, relative to the filesystems root. Thus in my > > exclude file I simply have "./dumps/amanda" listed and all 4 > > holding disk spaces are skipped. Perhaps you could comment > > out the 'holdingdisk no' parameter and replace it with an > > 'exclude file append ' for > > a night and see the effect. (check the syntax first) > > > > I'll try that. Thanks. > > What happens if I don't back up the holding disk ? I mean, on my > holding disk I have the dumps themselves, the logs, and the indexes. > > Should I exclude all this, or only the dumps themselves ? Just the dump space. Backup the config dir and other associated dirs. Some would actually reflect the previous run as they are not complete until the end of the dump. > If I lose the index data, it means I will no longer be able to use > amrecover right ? Correct, though amrestore or manual recovery will still work. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Can someone explain to me what no-hold in amstatus means ?
2005/9/14, Jon LaBadie <[EMAIL PROTECTED]>: > > An alternative, if you use gnutar rather than dump, is to exclude > the directory containing the holding disks space. I have my > holding disk spread over 4 drives. I named each consistantly > as dumps/amanda, relative to the filesystems root. Thus in my > exclude file I simply have "./dumps/amanda" listed and all 4 > holding disk spaces are skipped. Perhaps you could comment > out the 'holdingdisk no' parameter and replace it with an > 'exclude file append ' for > a night and see the effect. (check the syntax first) > I'll try that. Thanks. What happens if I don't back up the holding disk ? I mean, on my holding disk I have the dumps themselves, the logs, and the indexes. Should I exclude all this, or only the dumps themselves ? If I lose the index data, it means I will no longer be able to use amrecover right ?
Re: Can someone explain to me what no-hold in amstatus means ?
On Wed, Sep 14, 2005 at 10:30:16AM -0400, Guy Dallaire wrote: > 2005/9/14, Jon LaBadie <[EMAIL PROTECTED]>: > > > > I'm beginning to wonder if the reporting is faulty. I.e. the actual > > reason some dumpers is idle is not lack of holding space, but some > > idle time is mistakenly assigned to the no-hold category. > > > > I looked briefly at the source to driver.c. That is the only place > > I found that the string "no-hold" is used. It seems to me that a > > variable holding the reason for an idle dumper is not reset in a > > loop in driver.c. Thus, if some time there is a "no-hold" reason, > > perhaps the reason stays at "no-hold" for other reasons. > > > > There might be minor support for my theory if you have one or more > > DLEs that specifically invoke the 'holdingdisk no' parameter. For > > example, when the partition that contains the holdingdisk is backed > > up, often the use of the holdingdisk is prevented by the DLE. > > > > I'll post a not to the hackers list describing my theory. Maybe > > someone who understands the code can confirm or reject it. > > > > OK Thanks. I do have a single DLE that has the "holding disk no" > parameter, and that is my holding disk. I think you MUST specify > no-hold if you are backing up the holding disk. Yeah, there could be thrashing of the disk, and I can envision a situation that used to exist in old version of cat where it could continuously add to the end of a file: cat somefile >> somefile That situation is specifically tested for in cat now. A test whether that is the reason is to eliminate that parameter. The simplest is to just comment out that DLE in the disklist for a night and see the effect. But maybe other data on that DLE is too important for that. An alternative, if you use gnutar rather than dump, is to exclude the directory containing the holding disks space. I have my holding disk spread over 4 drives. I named each consistantly as dumps/amanda, relative to the filesystems root. Thus in my exclude file I simply have "./dumps/amanda" listed and all 4 holding disk spaces are skipped. Perhaps you could comment out the 'holdingdisk no' parameter and replace it with an 'exclude file append ' for a night and see the effect. (check the syntax first) jl -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Can someone explain to me what no-hold in amstatus means ?
2005/9/14, Jon LaBadie <[EMAIL PROTECTED]>: > > I'm beginning to wonder if the reporting is faulty. I.e. the actual > reason some dumpers is idle is not lack of holding space, but some > idle time is mistakenly assigned to the no-hold category. > > I looked briefly at the source to driver.c. That is the only place > I found that the string "no-hold" is used. It seems to me that a > variable holding the reason for an idle dumper is not reset in a > loop in driver.c. Thus, if some time there is a "no-hold" reason, > perhaps the reason stays at "no-hold" for other reasons. > > There might be minor support for my theory if you have one or more > DLEs that specifically invoke the 'holdingdisk no' parameter. For > example, when the partition that contains the holdingdisk is backed > up, often the use of the holdingdisk is prevented by the DLE. > > I'll post a not to the hackers list describing my theory. Maybe > someone who understands the code can confirm or reject it. > OK Thanks. I do have a single DLE that has the "holding disk no" parameter, and that is my holding disk. I think you MUST specify no-hold if you are backing up the holding disk.
Re: Can someone explain to me what no-hold in amstatus means ?
On Wed, Sep 14, 2005 at 09:01:46AM -0400, Guy Dallaire wrote: > Does it mean that there was not enough space in the holding disk and > amanda had to wait ? > > If so, why does amstatus tells me that my backup was in "no-hold" > about 95% of the time, when I backed up 8 gb and I had about 150 gb > free on the holding disk ? > > I've posted a question about that a couple of days ago but got now > explanation about the no-hold. > I'm beginning to wonder if the reporting is faulty. I.e. the actual reason some dumpers is idle is not lack of holding space, but some idle time is mistakenly assigned to the no-hold category. I looked briefly at the source to driver.c. That is the only place I found that the string "no-hold" is used. It seems to me that a variable holding the reason for an idle dumper is not reset in a loop in driver.c. Thus, if some time there is a "no-hold" reason, perhaps the reason stays at "no-hold" for other reasons. There might be minor support for my theory if you have one or more DLEs that specifically invoke the 'holdingdisk no' parameter. For example, when the partition that contains the holdingdisk is backed up, often the use of the holdingdisk is prevented by the DLE. I'll post a not to the hackers list describing my theory. Maybe someone who understands the code can confirm or reject it. jl -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Can someone explain to me what no-hold in amstatus means ?
Does it mean that there was not enough space in the holding disk and amanda had to wait ? If so, why does amstatus tells me that my backup was in "no-hold" about 95% of the time, when I backed up 8 gb and I had about 150 gb free on the holding disk ? I've posted a question about that a couple of days ago but got now explanation about the no-hold. Thanks a lot.
Re: Howto utilise DDS-4 20/40GB tape with Amanda
Hi PB My other issue is that concerns me If I also increase the number of tapes I currently use from 10 to possibly 20 will this enable me in the future to backup a server that is 80GB using incremental dumps and what is best for my monthly 20GB tapes that I wanted to pull out to archive as I wanted to do a full backup to 1 tape if this is now not possible could I still to the same as with my daily config and increase the runspercycle to 5 or more. I am using in my Daily amanda.conf dumpcycle 4 weeks # the number of days in the normal dump cycle # Setting dumpcycle to 0 - Setting this to zero tries to do a full backup each run. runspercycle 1 # the number of amdump runs in dumpcycle days # (4 weeks * 5 amdump runs per week -- just weekdays) tapecycle 10 tapes Would it be better to use your full suggestion as below: dumcycle 1 week runspercycle 5 # each working day tapecycle 20 tapes My Monthly is currently as below: dumpcycle 0 weeks # the number of days in the normal dump cycle runspercycle 1 # the number of amdump runs in dumpcycle days # (4 weeks * 5 amdump runs per week -- just weekdays) tapecycle 12 tapes But would these mean that I will have to change the tape frequently during the day. I am contemplating cannibalising a workstation and installing a DLT drive as a possible solution but cost rather a lot. Cheers On Mon, 2005-09-12 at 09:58 +0200, Paul Bijnens wrote: > Chuck Amadi Systems Administrator wrote: > > Hi relating to possible > > > > Here's my current amanda.conf > > > > dumpcycle 0 weeks > > runspercycle 1 > > tapecycle 10 tapes > > > > Thus increase runspercycle to any figure below tapecycle of 10 tapes. > > > > So for example runspercycle 1 would be sufficient. > > oho. You're doing a full dump each run. > But your tapecapacity is not sufficient to do that. > So that leaves you to do "runtapes 2" or more (use more > tapes for each run. > Or increase the dumpcycle (to avoid a full dump each time), and > then run multiple times during such a dumpcycle. > > Why not: > > dumcycle 1 week > runspercycle 5# each working day > tapecycle 10 tapes > > And if you need more history increase the number of tapes. > > > > > > Cheers > > > > > > > > > >>Chuck Amadi Systems Administrator wrote: > >> > >>>Hi I have managed to config the hard disk partitions in my disklist but > >>>I am short of tape space to include the remainder of the hard disk > >>>partitions. > >>> > >>>I had initially run the following commands is it possible to get > >>>compression functioning to enable to use 40GB as opposed to native 20GB. > >>> > >>>* - amtapetype -f /dev/nst0 # determine tape type takes along time > >> > >>When specifying also a realistic estimate "-e 20 g" it is usually much > >>faster (2-4 hours) > >> > >> > >> > >>>* - turn off datacompression run mt command that controls magnetic tape > >>>drive operation. * - mt -f /dev/nst0 status and mt -f /dev/nst0 > >>>datcompression 0 Compression off. > >>> > >>>As I don't want to buy bigger tapes if possible. > >> > >> > >>If you already use software compression, than enabling hardware > >>compression will _reduce_ your effective tapecapacity by 15-35%, because > >>the compression algorithms in those tapedrives behave bad on already > >>compressed data. > >> > >>When not too many files change, you can run less fulls and more > >>incrementals on each run by increasing the dumpcycle (runspercycle > >>actually). Or increase "runtapes" (and flush each morning the > >>remainders if you have only one drive). > >> > >> > > -- Unix/ Linux Systems Administrator Chuck Amadi The Surgical Material Testing Laboratory (SMTL), Princess of Wales Hospital Coity Road Bridgend, United Kingdom, CF31 1RQ. Email chuck.smtl.co.uk Tel: +44 1656 752820 Fax: +44 1656 752830
