Re: amtape DailySet1 show --- asks for slot 2 ??
On Thu, Jul 13, 2006 at 08:19:12PM -0700, Brian Neu wrote: I'm confused. Why does this command eject the tape and then ask for me to insert a tape into slot 2? -sh-3.1$ amtape DailySet1 show changer: got exit: 0 str: 1 99 1 changer_query: changer return was 99 1 changer_query: searchable = 0 amtape: scanning all 99 slots in tape-changer rack: changer_find: looking for NULL changer is searchable = 0 changer: got exit: 0 str: 1 /dev/nst0 slot 1: date Xlabel DailySet1-04 insert tape into slot 2 and press return It appears you are using the chg-manual tape changer script. I'm guessing it is configured for 99 slots. With the show command of amtape you have asked it to tell you which tape is in each of the 99 slots. As the changer is manual who, except you, is supposed to change the tapes? -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Reset By Peer
On 2006-07-13 20:00, James Wilson wrote: Has anyone gotten these errors before and if so do you know why it happens? write_data: write failure. Error = Connection reset by peer ? write_socket: Error writing 39 bytes to socket 11: ERRNO = Connection reset by peer ? Error writing 39 bytes to client. -1 (Connection reset by peer) sendbackup: size 111879101 sendbackup: end Maybe it is an instance of the problem described here: http://wiki.zmanda.com/index.php/Amdump:_mesg_read:_Connection_reset_by_peer Otherwise we'll need more info. -- Paul Bijnens, xplanation Technology ServicesTel +32 16 397.511 Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax +32 16 397.512 http://www.xplanation.com/ email: [EMAIL PROTECTED] *** * I think I've got the hang of it now: exit, ^D, ^C, ^\, ^Z, ^Q, ^^, * * F6, quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, * * stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt, abort, hangup, * * PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e, kill -1 $$, shutdown, * * init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... * * ... Are you sure? ... YES ... Phew ... I'm out * ***
Re: To whom it may concern
Jon LaBadie schrieb: On Fri, Jul 14, 2006 at 01:37:53AM +0200, Stefan G. Weichinger wrote: My best greetings to all of you, the amanda-users and -hackers, ... But I am also accepting the fact, that all this is not MINE and that I have to let it go as well. All the best to all of you, Must have been tough to write your message. Not really, I just waited for the right point in time. I just wanted to get that load off and tell the community. The best to you in whatever ventures you persue. I hope we continue to meet. Sure we will. As I pointed out I am not quitting the team, I am just officially going into passive mode. Maybe I catch my breath then ... Thanks, Stefan
Re: Dealing more gracefully with tape overruns
On Wed, Jul 12, 2006 at 04:30:12PM +0200, Ronan KERYELL wrote: If not, what about adding a new strategy with a mode such as: dump-on-holding-disk-if-degraded-mode yes Should not be to difficult to implement. For this there is the maxdumpsize 1g, together with reserve 0. (use appropriate numbers for your config). I will try also to play with the reserve keyword next night with, for example, reserve 50 % for full backups. reserve number Default: 100. The part of holding-disk space that should be reserved for incremental backups if no tape is available, expressed as a percentage of the available holding-disk space (0-100). By default, when there is no tape to write to, degraded mode (incremental) backups will be performed to the holding disk. If full backups should also be allowed in this case, the amount of holding disk space reserved for incrementals should be lowered. But if it works, why this static allocation between incremental and full backups? Note that the reserve parameter comes only into play when there is tape problem. There is no static allocation. In the planning phase, Amanda creates a schedule to dump DLE's that is limited by the amount specified in maxdumpsize or available tape, if omited. It also generates a plan B for use by the driver, that is used when there suddenly is a tapeproblem. When the driver notices there is a tape problem, it switches to degraded mode and uses that plan B: it will replace some full dumps with incrementals instead. Only in that case, a part of the holdingdisk is reserved for incrementals only (specified by the reserve parameter). Another approach: what about also a mode with runtapes 1 without any changer, assuming that the user will use amflush later? I have run such a config for a few years, until I bought a bigger tape drive. I had a config with two external tapes, configured as a changer with chg-multi. The archive backups in the weekend actually needed 4 tapes (sometimes even 5). I had just set maxdumpsize equivalent to the volume of 5 tapes and had reserve 0 (nothing reserved for incrementals only), and a big holdingdisk. Two tapes were online, and were filled during the run. The rest of the data was kept on the large holdingdisk, and when I came in on monday morning, I just inserted the next two tapes and ran amflush. Sometimes I had to flush even once again to a 5 or 6th tape (the tapedrives were aging, and gave IO-errors now and then). -- Paul Bijnens, xplanation Technology ServicesTel +32 16 397.511 Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax +32 16 397.512 http://www.xplanation.com/ email: [EMAIL PROTECTED] *** * I think I've got the hang of it now: exit, ^D, ^C, ^\, ^Z, ^Q, ^^, * * F6, quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, * * stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt, abort, hangup, * * PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e, kill -1 $$, shutdown, * * init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... * * ... Are you sure? ... YES ... Phew ... I'm out * ***
Re: Backup for specific files
On 2006-07-13 15:52, silpa kala wrote:
Hi,
I am trying to take the backup for
.log/.txt/executable file name. But the planner is not
able to provide the estimations.
I observed that only we can take the backup for the
directories. Why this one is happening?
Please provide the details is there any restrictions
to provide to take the backup for the file type?
Amanda uses (currently) two external program to do the actual
backup: dump (several flavours actually), and gnutar.
As an intelligent person, you can quickly hunt through the source
tree and find the places where those programs are executed.
There you see that the several flavours of dump actually
need (device name of) a partition. So here it is impossible
to backup one file (or even part of partition).
The place where gnutar is invoked is also easily spotted, and
you'll see that the command line is assembled as:
gnutar ... --directory /the/dle/name ...
And then you understand that /the/dle/name needs to be a
directory, and not a file.
But if you want to backup one file, there is a simple solution
too, by using a DLE like:
client.example.com /path/to/file /path/to {
comp-user-tar
include ./file
}
Using the manpages, you'll understand what the above syntax
is all about, and seeing the source code, you'll also understand
how that works to accomplish what you're trying to do.
--
Paul Bijnens, xplanation Technology ServicesTel +32 16 397.511
Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax +32 16 397.512
http://www.xplanation.com/ email: [EMAIL PROTECTED]
***
* I think I've got the hang of it now: exit, ^D, ^C, ^\, ^Z, ^Q, ^^, *
* F6, quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, *
* stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt, abort, hangup, *
* PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e, kill -1 $$, shutdown, *
* init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... *
* ... Are you sure? ... YES ... Phew ... I'm out *
***
Re: Exclude list entries
Thanks for your replies. I could have phrased my question slightly better. I want to know what the exclude list should look like to make sure I exclude all files with these extensions in any directory/subdirectory contained under the disklist entries: .ora .dbf .dmp .dmp.gz.xx If I understand you correctly, then using the following will only exclude files with those extensions located within top level directories, and not necessarily subdirectories, e.g. ./*.ora Is that correct? Olivier Nicole wrote: No there really is a difference between excluding /foo, ./foo, and foo. As you are backing up ., /foo will not match anything. Of course ./foo will match any foo in the top level directory . foo will match any foo in any directory under . Not so sure: banyanon: /usr/local/bin/tar --version tar (GNU tar) 1.13.25 Copyright (C) 2001 Free Software Foundation, Inc. This program comes with NO WARRANTY, to the extent permitted by law. You may redistribute it under the terms of the GNU General Public License; see the file named COPYING for details. Written by John Gilmore and Jay Fenlason. banyanon: /usr/local/bin/tar cfv /dev/null . | grep jpg ./Images/Picture-2 003.jpg ./Images/Picture-2 005.jpg ./maman.jpg banyanon: cat ../excl ./*.jpg banyanon: /usr/local/bin/tar cfv /dev/null --exclude-from=../excl . | grep jpg banyanon: cat ../excl2 *.jpg banyanon: /usr/local/bin/tar cfv /dev/null --exclude-from=../excl2 . | grep jpg Excl contains ./*.jpg and excl2 contains *.jpg but they both exclude any .jpg file in any level of the hierarchy. OK, funny enough, tar 1.13.19 on another machine gives different results, but still, tar 1.13.25 is recommeneded version isn't it? Olivier -- View this message in context: http://www.nabble.com/Exclude-list-entries-tf1936459.html#a5323781 Sent from the Amanda - Users forum at Nabble.com.
Re: Exclude list entries
I want to know what the exclude list should look like to make sure I exclude all files with these extensions in any directory/subdirectory contained under the disklist entries: .ora .dbf .dmp .dmp.gz.xx It seems that the safest way is: *.ora *.dbf *.dmp *.dmp.gz.* without the initial ./ Olivier
Re: Partial backup
What I meant before the backup is that I ran amdump manually waited for about a minute because it does the estimated size, then I ran amstatus. Then I ran amstatus again when it finished I do have the email after the backup if you want to see that. But there should be at least 2 gigs of data if not more that it should be backing up from that host. I am using an SMB share to back this host up. And I get a lot of timeout errors maybe you have seen this before? I'll paste the output from the email below. FAILED AND STRANGE DUMP DETAILS: /-- amanda.transolutions.net //192.168.1.133/ifxsessions lev 1 STRANGE sendbackup: start [amanda.transolutions.net://192.168.1.133/ifxsessions level 1] sendbackup: info BACKUP=/usr/bin/smbclient sendbackup: info RECOVER_CMD=/bin/gzip -dc |/usr/bin/smbclient -f... - sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end | Domain=[IFX-SE-02] OS=[Windows 5.0] Server=[Windows 2000 LAN Manager] ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\20060501\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2
Re: Partial backup
On 2006-07-14 15:13, James Wilson wrote: What I meant before the backup is that I ran amdump manually waited for about a minute because it does the estimated size, then I ran amstatus. Then I ran amstatus again when it finished I do have the email after the backup if you want to see that. But there should be at least 2 gigs of data if not more that it should be backing up from that host. I am using an SMB share to back this host up. And I get a lot of timeout errors maybe you have seen this before? I'll paste the output from the email below. There you have it: The estimate was easy, and said it would be about X Gbyte. But doing the real backup, there were problems: timeout errors, and so that what resulted as backup was much less then promised. Now just finding out what is the cause of those timeout errors. FAILED AND STRANGE DUMP DETAILS: /-- amanda.transolutions.net //192.168.1.133/ifxsessions lev 1 STRANGE sendbackup: start [amanda.transolutions.net://192.168.1.133/ifxsessions level 1] sendbackup: info BACKUP=/usr/bin/smbclient sendbackup: info RECOVER_CMD=/bin/gzip -dc |/usr/bin/smbclient -f... - sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end | Domain=[IFX-SE-02] OS=[Windows 5.0] Server=[Windows 2000 LAN Manager] ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\20060501\* And it seems to be smbclient problem. [...] | tar: dumped 178 files and directories | Total bytes written: 11032576 sendbackup: size 10774 sendbackup: end \ -- Paul Bijnens, xplanation Technology ServicesTel +32 16 397.511 Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax +32 16 397.512 http://www.xplanation.com/ email: [EMAIL PROTECTED] *** * I think I've got the hang of it now: exit, ^D, ^C, ^\, ^Z, ^Q, ^^, * * F6, quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, * * stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt, abort, hangup, * * PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e, kill -1 $$, shutdown, * * init 0, kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ... * * ... Are you sure? ... YES ... Phew ... I'm out * ***
Release of amanda-2.5.1b1
Hello, The Amanda core team is pleased to announce the first beta release of Amanda 2.5.1, the 2.5.1b1 release. It can be dowloaded from http://www.amanda.org (not yet available) or https://sourceforge.net/project/showfiles.php?group_id=120 Here's a list of the changes for release 2.5.1b1 (from the NEWS file): Look at the ReleaseNotes and ChangeLog file for more details. * work with tar-1.15.91 * Open SSL encryption support * amfetchdump -o is replaced by -O. * amrecover_do_fsf default to yes * amrecover_check_label default to yes * amcheck -w will do all test including the tape writable test. use amcheck -t -w if you want only the tape writable test. * usetimestamps * bsdudp/bsdtcp, 2 new auth. * Unlimited number of DLE on a client with bsdtcp/rsh or ssh auth. * docs/howto-auth.txt * amandad_path, client_username and ssh_keys (for ssh/rsh) * client config file (see man amanda-client.conf) Only use by amrecover. * amrecover use the security-api * amoldrecover which use the old amindexd/amidxtaped protocol * -o option to overwrite config option (see in man amanda, the CONFIGURATION OVERWRITE section) If you find bug in this release, please submit a bug report on sourceforge in the group v2.5.1b1. (http://sourceforge.net/tracker/?group_id=120atid=100120)
Re: To whom it may concern
Thanks Stefan for all the good work !
Re: Exclude list entries
On Fri, Jul 14, 2006 at 04:41:37PM +0700, Olivier Nicole wrote: I want to know what the exclude list should look like to make sure I exclude all files with these extensions in any directory/subdirectory contained under the disklist entries: .ora .dbf .dmp .dmp.gz.xx It seems that the safest way is: *.ora *.dbf *.dmp *.dmp.gz.* without the initial ./ And to confirm it will work, put your test patterns in a file, say exclude-pats. Then cd to the directory of your DLE. Run something like: tar -cvf /dev/null --exclude-from exclude-pats 21 | egrep 'ora|dbf|dmp' There should be nothing in the output that matches your list. Might be something like oracle.txt, but that shouldn't be excluded. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Partial backup
On Fri, Jul 14, 2006 at 08:13:40AM -0500, James Wilson wrote: body bgcolor=#ff text=#00 What I meant before the backup is that I ran amdump manually waited for about a minute because it does the estimated size, then I ran amstatus. Then I ran amstatus again when it finished I do have the email after the backup if you want to see that. But there should be at least 2 gigs of data if not more that it should be backing up from that host. I am using an SMB share to back this host up. And I get a lot of timeout errors maybe you have seen this before? I'll paste the output from the email below.br br pre wrap=FAILED AND STRANGE DUMP DETAILS: /-- amanda.transolutions.net //192.168.1.133/ifxsessions lev 1 STRANGE sendbackup: start [amanda.transolutions.net://192.168.1.133/ifxsessions level 1] sendbackup: info BACKUP=/usr/bin/smbclient sendbackup: info RECOVER_CMD=/bin/gzip -dc |/usr/bin/smbclient -f... - sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end | Domain=[IFX-SE-02] OS=[Windows 5.0] Server=[Windows 2000 LAN Manager] ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\20060501\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* ? Call timed out: server did not respond after 2 milliseconds listing \Apex_SPFLD_ADT\Input\Archive\2\* As the first status was early in the same run as the second status, I'll change my thoughts. The two reports clearly said different size estimates, 1231m vs 171m. But the first said partial estimate done. Amanda typically does estimates for a full dump and one or more incremental levels. After that the decision as to what level backup to actually perform is done. My guess is that your first status report caught amanda after the level 0 estimate (1231m) and its decision to do an incremental dump (171m). Amanda appears to have chosen to do the smaller incremental dump. Your amreport would allow you to confirm this: HOSTNAME DISKL ORIG-MB OUT-MB ... -- -- ... bigcow HOME1 227 12 ... bigcow LOCAL 1 125122 ... bigcow LastChanceC 0 32 17 ... The L column is the level. As to why you got only 5m out of 171m expected, well ... it is a PC ;( Note my last line above. LastChance is my wife's PC and the trailing 'C' in the disk name refers of course to the C drive. Only 32 MB were dumped on a level 0. This is from an old run, I now dump over 7 GB on a level 0. I too was, and to a lesser degree still am getting those timed out messages. I've not seen them mentioned on the list, but I have basically said - 'thats samba backups of a PC for yah'. You can more easily do that in a home rather than in a business environment. Particularly if it is not your data ;) But I did go from 32 MB to 7 GB with permission changes. When I tried to make changes from a normal boot up, I was unable to accomplish much. Then I read in some Micro$oft note that somethings can only be done by the actual administrator account, not by some other user with adminis- trator privilege. So I rebooted the PC into safe-mode (the only way to get to the actual administrator account I read). Then, on each drive, I set the user amanda accesses to have full control, not just read, and clicked the button to propogate the change down to all files and folders. Suddenly I was backing up much more data and incrementals were smaller than full dumps. I don't recommend the changes I did. Not being a PC person I don't know all the consequences. And it might be a policy violation in a business. But it made a big difference for me. There are still lots of files that I get errors about not being able to be backed up. And I still get those timed out messages. But now I'm getting the bulk of the data and it appears to include most (all?) of what I'm looking to backup. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Re: Exclude list entries
On Fri, Jul 14, 2006 at 10:54:07AM +0700, Olivier Nicole wrote: No there really is a difference between excluding /foo, ./foo, and foo. As you are backing up ., /foo will not match anything. Of course ./foo will match any foo in the top level directory . foo will match any foo in any directory under . Not so sure: Gee, can we be sure of anything with gnutar ;) banyanon: /usr/local/bin/tar --version tar (GNU tar) 1.13.25 Copyright (C) 2001 Free Software Foundation, Inc. This program comes with NO WARRANTY, to the extent permitted by law. You may redistribute it under the terms of the GNU General Public License; see the file named COPYING for details. Written by John Gilmore and Jay Fenlason. banyanon: /usr/local/bin/tar cfv /dev/null . | grep jpg ./Images/Picture-2 003.jpg ./Images/Picture-2 005.jpg ./maman.jpg banyanon: cat ../excl ./*.jpg banyanon: /usr/local/bin/tar cfv /dev/null --exclude-from=../excl . | grep jpg banyanon: cat ../excl2 *.jpg banyanon: /usr/local/bin/tar cfv /dev/null --exclude-from=../excl2 . | grep jpg Excl contains ./*.jpg and excl2 contains *.jpg but they both exclude any .jpg file in any level of the hierarchy. OK, funny enough, tar 1.13.19 on another machine gives different results, but still, tar 1.13.25 is recommeneded version isn't it? I thought either. Fessing up, my test was with tar 1.15.1 and actually I used --exclude 'pattern' rather than --exclude-from file_of_patterns. Beginning to sound like its try til you get what works for you. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
RE: bygwin amanda client behind firewall.
Well I am getting a little closer. My cygwin client is a Window2003 server. When I do a amcheck from the amanda server to the cygwin client behind the firewall I now get the following in the events log on the cygwin client. The following information is part of the event: inetd: PID 2280: amanda: can't set uid 18: Permission denied. ...The following information is part of the event: inetd: PID 2764: /local/Amanda/amanda-2.4.5p1/libexec/amandad: exit status 0x100. So now I am at least getting to the inetd service. The following are some things that I have done to tray an track down why amanda can't set uid 18: 1) I have run /local/Amanda/amanda-2.4.5p1/libexec/amandad manually and it times out after 30 seconds. From the amandad.debug when run manually. amandad: time 30.000: dgram_recv: timeout after 30 seconds amandad: error receiving message: timeout amandad: time 30.000: error receiving message: timeout amandad: time 30.000: pid 2276 finish time Fri Jul 14 10:31:47 2006 2) I have compared my Window2003 system with a working WinXP client that works. I compared the /tmp/amanda, /etc/amandate, /local//var/gunlist-tar, services, ineted.conf, .amandahosts, /etc/passwd, to make sure the permissions and owner are the same. I can telnet from the server to the client and I can make a connection so I believe inetd is work. 3) I have looked in the /var/log/inetd.log. No messages. 4) There are no messages in /tmp/amanda. 5) I have recompiled amanda. 6) This is my inetd.conf amanda entry. amanda dgram udp waitSYSTEM /local/Amanda/amanda-2.4.5p1/libexec/amandadamandadDo you see anything wrong. I realize that this is probably a cygwin news group question but just want to ask it anyone else had this problem on a Window2003 client. Do you have any other ideas on where to look for error messages or how to glean error messages from this run? _ Robert P. McGraw, Jr. Manager, Computer System EMAIL: [EMAIL PROTECTED] Purdue University ROOM: MATH-807 Department of MathematicsPHONE: (765) 494-6055 150 N. University Street FAX: (419) 821-0540 West Lafayette, IN 47907-2067 smime.p7s Description: S/MIME cryptographic signature
Re: bygwin amanda client behind firewall.
On Fri, Jul 14, 2006 at 03:10:34PM -0400, McGraw, Robert P. wrote: Well I am getting a little closer. My cygwin client is a Window2003 server. When I do a amcheck from the amanda server to the cygwin client behind the firewall I now get the following in the events log on the cygwin client. The following information is part of the event: inetd: PID 2280: amanda: can't set uid 18: Permission denied. ...The following information is part of the event: inetd: PID 2764: /local/Amanda/amanda-2.4.5p1/libexec/amandad: exit status 0x100. So now I am at least getting to the inetd service. ... 6) This is my inetd.conf amanda entry. amanda dgram udp wait SYSTEM /local/Amanda/amanda-2.4.5p1/libexec/amandadamandad Do you see anything wrong. Is the amanda-user SYSTEM (uid = 18). I forget how it needs to be on cygwin clients, but on unix amandad would be started as the amanda-user, not as root. Probably just a fluke of mailers, but in my mailreader the two amandad's are run together with no space as in: .../libexec/amandadamandad -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
RE: bygwin amanda client behind firewall.
-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jon LaBadie Sent: Friday, July 14, 2006 3:29 PM To: amanda-users@amanda.org Subject: Re: bygwin amanda client behind firewall. On Fri, Jul 14, 2006 at 03:10:34PM -0400, McGraw, Robert P. wrote: Well I am getting a little closer. My cygwin client is a Window2003 server. When I do a amcheck from the amanda server to the cygwin client behind the firewall I now get the following in the events log on the cygwin client. The following information is part of the event: inetd: PID 2280: amanda: can't set uid 18: Permission denied. ...The following information is part of the event: inetd: PID 2764: /local/Amanda/amanda-2.4.5p1/libexec/amandad: exit status 0x100. So now I am at least getting to the inetd service. ... 6) This is my inetd.conf amanda entry. amanda dgramudp wait SYSTEM /local/Amanda/amanda- 2.4.5p1/libexec/amandadamandad Do you see anything wrong. Is the amanda-user SYSTEM (uid = 18). [McGraw, Robert P.] In my /etc/passwd I have the following: SYSTEM:*:18:18:,S-1-5-18:/home/root: I forget how it needs to be on cygwin clients, but on unix amandad would be started as the amanda-user, not as root. [McGraw, Robert P.] Per the amanda docs for cygwin clients, it is the SYSTEM account that it runs under. Probably just a fluke of mailers, but in my mailreader the two amandad's are run together with no space as in: .../libexec/amandadamandad [McGraw, Robert P.] Yeah I saw this right after I sent it out. It is amandad amandad I was cut and pasting to get it into the email and must have lost the tab. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road(609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax) [McGraw, Robert P.] I have checked, and rechecked for spelling etc. I have also compared my WinXP cygwin client, which works with the Win2003 client. I have send email to the cygwin news group asking for help. Thanks Robert smime.p7s Description: S/MIME cryptographic signature
Re: bygwin amanda client behind firewall.
McGraw, Robert P. wrote: Well I am getting a little closer. My cygwin client is a Window2003 server. So is mine. When I do a amcheck from the amanda server to the cygwin client behind the firewall I now get the following in the events log on the cygwin client. The following information is part of the event: inetd: PID 2280: amanda: can't set uid 18: Permission denied. ...The following information is part of the event: inetd: PID 2764: /local/Amanda/amanda-2.4.5p1/libexec/amandad: exit status 0x100. So now I am at least getting to the inetd service. The following are some things that I have done to tray an track down why amanda can't set uid 18: Documentation says that you should use tab and not spaces to separate the tokens on inetd.conf. I don't know if that is really an issue but it's always better to remember. And you can check which user inetd is running as. To find that, dclick the taskbar, Task manager, go to the Processes tab, select the Show all users processes checkbox, find inetd and check which user it's running as. It should be running as SYSTEM. If it's not, that could be the problem. I would add that I was running as user Administrator of the local machine (not administrator of the domain) when I ran inetd with --install-as-service. Bye! Robert
