Solaris 11.4 issues for Amanda 3.6.0.pre2
Hi I'm testing the release on Solaris 11.4 and the good news is that autogen/configure and build works with the gcc tools provided with the system it builds the 64 binaries with native Perl. Only one Perl module was missing Encode-Locale. Still there is an issues with ambind se debug listings below. I can provide more truss debugging if needed. I guess it has never been tested on Solaris 11 ? Thanks GG Amrecover Fri May 27 18:19:56.346914960 2022: pid 7819: amrecover- [..d1ed]: security_getdriver(name=bsdtcp) returns 77994368 Fri May 27 18:19:56.347518100 2022: pid 7819: amrecover- [..d7e9]: security_handleinit(handle=10017d5e0, driver=77994368 (BSDTCP)) Fri May 27 18:19:56.348500480 2022: pid 7819: amrecover- [..d7e9]: security_streaminit(stream=1001804d0, driver=77994368 (BSDTCP)) Fri May 27 18:19:56.348663580 2022: pid 7819: amrecover- [..d7e9]: connect_port: Skip port 512: owned by exec. Fri May 27 18:19:56.348716800 2022: pid 7819: amrecover- [..d7e9]: connect_port: Skip port 513: owned by login. Fri May 27 18:19:56.348766780 2022: pid 7819: amrecover- [..d7e9]: connect_port: Skip port 514: owned by shell. Fri May 27 18:19:56.348816170 2022: pid 7819: amrecover- [..d7e9]: connect_port: Skip port 515: owned by printer. Fri May 27 18:19:56.348975230 2022: pid 7819: amrecover- [..d7e9]: make_socket opening socket with family 2 Fri May 27 18:19:56.349004870 2022: pid 7819: amrecover- [..d7e9]: make_socket opening socket with family 2: 4 Fri May 27 18:19:56.350090030 2022: pid 7819: amrecover- [..d7e9]: ambind failed: sendmsg failed A: Message too long Fri May 27 18:19:56.350158680 2022: pid 7819: amrecover- [..d7e9]: stream_client: Could not bind to port in range 512-1023. Fri May 27 18:19:56.350193430 2022: pid 7819: amrecover- [..d7e9]: security_seterror(handle=10017d5e0, driver=77994368 (BSDTCP) error=sendmsg failed A: Message too long amcheck -c daily iodomd / Amanda Backup Client Hosts Check ERROR: iodomd: selfcheck request failed: sendmsg failed A: Message too long Client check: 1 host checked in 13.007 seconds. 1 problem found. (brought to you by) Fri May 27 20:47:26.760841910 2022: pid 9701: amcheck-clients- [..edef]: security_getdriver(name=bsdtcp) returns 77994368 Fri May 27 20:47:26.761782860 2022: pid 9701: amcheck-clients- [..b3ba]: security_handleinit(handle=10046c700, driver=77994368 (BSDTCP)) Fri May 27 20:47:26.762759830 2022: pid 9701: amcheck-clients- [..b3ba]: security_streaminit(stream=100471be0, driver=77994368 (BSDTCP)) Fri May 27 20:47:26.762905150 2022: pid 9701: amcheck-clients- [..b3ba]: connect_port: Skip port 512: owned by exec. Fri May 27 20:47:26.762956770 2022: pid 9701: amcheck-clients- [..b3ba]: connect_port: Skip port 513: owned by login. Fri May 27 20:47:26.763006350 2022: pid 9701: amcheck-clients- [..b3ba]: connect_port: Skip port 514: owned by shell. Fri May 27 20:47:26.763055780 2022: pid 9701: amcheck-clients- [..b3ba]: connect_port: Skip port 515: owned by printer. Fri May 27 20:47:26.763219520 2022: pid 9701: amcheck-clients- [..b3ba]: make_socket opening socket with family 2 Fri May 27 20:47:26.763250810 2022: pid 9701: amcheck-clients- [..b3ba]: make_socket opening socket with family 2: 4 Fri May 27 20:47:26.764355740 2022: pid 9701: amcheck-clients- [..b3ba]: ambind failed: sendmsg failed A: Message too long Fri May 27 20:47:26.764433720 2022: pid 9701: amcheck-clients- [..b3ba]: stream_client: Could not bind to port in range 512-1023. Fri May 27 20:47:26.764510810 2022: pid 9701: amcheck-clients- [..b3ba]: security_seterror(handle=10046c700, driver=77994368 (BSDTCP) error=sendmsg failed A: Message too long
Re: Sent out a pull-request to upgrade the packaging system....
Is it a problem with the Solaris port for the 4.5.* version related ambind ? We have never seen this problem in older versions. Let me know if more debugging is needed. Have you been able to reproduce the error ? Thanks Gunnar Skickat från min iPhone > 23 mars 2022 kl. 17:42 skrev Chris Hassell : > > My guess from the manpage of sendmsg() is that this applies. The > SOCK_DGRAM may not working as a transport. > > to specify the size. The length of the message is supplied by the len > parameter. For socket types such as SOCK_DGRAM and SOCK_RAW that > require atomic messages, the error EMSGSIZE is returned and the message > is not transmitted when it is too long to pass atomically through the > underlying protocol. The same restrictions do not apply to SOCK_STREAM > sockets. > > So as we never use or verify the datagram protocol ourselves. I have > wondered how fast it streams from a big client. > > I am guessing here that some problem shows up whence we're sending longer > packets than the 1460 or so that can be supported? > >> -Original Message- >> From: Gunnarsson, Gunnar >> Sent: Tuesday, March 22, 2022 9:30 AM >> To: Chris Hassell ; li...@xunil.at; ? amanda users >> >> Subject: SV: Sent out a pull-request to upgrade the packaging system >> >> WARNING: This email originated from outside of BETSOL. Do not click links or >> open attachments unless you recognize the sender and know the content is >> safe. >> >> >> Hi, >> >> I'm testing the amanda-3_5-packaging-upgrade-wip and now it builds on my >> solaris 11.4 system. >> Amrecover fails to connect to the amindexd server see trace from below. >> Complains about too big message size see trace below. >> Server works if the amrecover client is 3.4.5 and the backup client seems to >> be >> working using th 3.4.5 version. >> >> Thanks Gunnar >> >> Tue Mar 22 15:42:21.445289090 2022: pid 11472: thd-100124a00: amindexd: pid >> 11472 ruid 91 euid 91 version 3.5.1+233: start at Tue Mar 22 15:4 >> 2:21 2022 >> Tue Mar 22 15:42:21.445353790 2022: pid 11472: thd-100124a00: amindexd: >> version 3.5.1+233 >> >> 11484/2:setsockopt(4, SOL_SOCKET, SO_REUSEADDR, >> 0x7E8FB07C, 4, SOV_DEFAULT) = 0 >> 11484/2:setsockopt(4, SOL_SOCKET, SO_KEEPALIVE, 0x7E8FB07C, >> 4, SOV_DEFAULT) = 0 >> 11484/2:so_socket(PF_UNIX, SOCK_DGRAM|SOCK_NONBLOCK, 0, NULL, >> SOV_XPG4_2) = 5 >> 11484/2:so_socket(PF_UNIX, SOCK_DGRAM|SOCK_NONBLOCK, 0, NULL, >> SOV_XPG4_2) = 6 >> 11484/2:so_socketpair(5, 6) = 0 >> 11484/2:pipe() = 7 [8] >> 11484/2:lwp_suspend(1) = 0 >> 11484/2:forkx(0)= 11487 >> 11487: forkx() (returning as child of pid 11484) = 0 >> 11484/2:lwp_continue(1) = 0 >> 11487: getpid()= 11487 [11484] >> 11484/2:lwp_sigmask(SIG_SETMASK, 0x, 0x, >> 0x, 0x) = 0xFFBFFEFF [0x] >> 11487: lwp_self() = 2 >> 11484/2:close(6)= 0 >> 11484/2:close(8)= 0 >> 11487: lwp_sigmask(SIG_SETMASK, 0x, 0x, 0x, >> 0x) = 0xFFBFFEFF [0x] >> 11484/2:sendmsg(5, 0x7E8F8E48, MSG_XPG4_2) Err#97 >> EMSGSIZE >> >> ^ >> ^^^ >> 11487: schedctl() = 0x7F5E4000 >> 11484/2:shutdown(5, SHUT_RDWR, SOV_DEFAULT) = 0 >> 11484/2:close(5)= 0 >> 11484/2:close(7)= 0 >> 11487: close(5)= 0 >> 11484/2:close(4)= 0 >> 11484/2:getpid()= 11484 >> [11483] >> 11487: fcntl(8, F_DUP2FD, 0x0002) = 2 >> 11484/2:write(10, 0x100147A64, 124) = 124 >> 11484/2: r e c o v e r : a m b i n d f a i l e d : s e n d m >> s g >> 11484/2: f a i l e d A : M e s s a g e t o o l o n g\n\n >> 11487: fcntl(0, F
SV: Sent out a pull-request to upgrade the packaging system....
Hi, I'm testing the amanda-3_5-packaging-upgrade-wip and now it builds on my solaris 11.4 system. Amrecover fails to connect to the amindexd server see trace from below. Complains about too big message size see trace below. Server works if the amrecover client is 3.4.5 and the backup client seems to be working using th 3.4.5 version. Thanks Gunnar Tue Mar 22 15:42:21.445289090 2022: pid 11472: thd-100124a00: amindexd: pid 11472 ruid 91 euid 91 version 3.5.1+233: start at Tue Mar 22 15:4 2:21 2022 Tue Mar 22 15:42:21.445353790 2022: pid 11472: thd-100124a00: amindexd: version 3.5.1+233 11484/2:setsockopt(4, SOL_SOCKET, SO_REUSEADDR, 0x7E8FB07C, 4, SOV_DEFAULT) = 0 11484/2:setsockopt(4, SOL_SOCKET, SO_KEEPALIVE, 0x7E8FB07C, 4, SOV_DEFAULT) = 0 11484/2:so_socket(PF_UNIX, SOCK_DGRAM|SOCK_NONBLOCK, 0, NULL, SOV_XPG4_2) = 5 11484/2:so_socket(PF_UNIX, SOCK_DGRAM|SOCK_NONBLOCK, 0, NULL, SOV_XPG4_2) = 6 11484/2:so_socketpair(5, 6) = 0 11484/2:pipe() = 7 [8] 11484/2:lwp_suspend(1) = 0 11484/2:forkx(0)= 11487 11487: forkx() (returning as child of pid 11484) = 0 11484/2:lwp_continue(1) = 0 11487: getpid()= 11487 [11484] 11484/2:lwp_sigmask(SIG_SETMASK, 0x, 0x, 0x, 0x) = 0xFFBFFEFF [0x] 11487: lwp_self() = 2 11484/2:close(6)= 0 11484/2:close(8)= 0 11487: lwp_sigmask(SIG_SETMASK, 0x, 0x, 0x, 0x) = 0xFFBFFEFF [0x] 11484/2:sendmsg(5, 0x7E8F8E48, MSG_XPG4_2) Err#97 EMSGSIZE 11487: schedctl() = 0x7F5E4000 11484/2:shutdown(5, SHUT_RDWR, SOV_DEFAULT) = 0 11484/2:close(5)= 0 11484/2:close(7)= 0 11487: close(5)= 0 11484/2:close(4)= 0 11484/2:getpid()= 11484 [11483] 11487: fcntl(8, F_DUP2FD, 0x0002) = 2 11484/2:write(10, 0x100147A64, 124) = 124 11484/2: r e c o v e r : a m b i n d f a i l e d : s e n d m s g 11484/2: f a i l e d A : M e s s a g e t o o l o n g\n\n 11487: fcntl(0, F_GETFD) = 0 11484/2:getpid()= 11484 [11483] 11487: fcntl(1, F_GETFD) = 0 11484/2:write(10, 0x100147A64, 130) = 130 11484/2: T u e M a r 2 2 1 5 : 4 4 : 5 4 . 0 4 9 9 5 3 7 3 0 2 0 11484/2: 2 2 : p i d 1 1 4 8 4 : t h d - 1 0 0 1 5 5 0 0 0 : a m 11484/2: r e c o v e r : s t r e a m _ c l i e n t : C o u l d n o 11484/2: t b i n d t o p o r t i n r a n g e 5 1 2 - 1 0 2 3 11484/2: .\n 11487: fcntl(2, F_GETFD) = 0 11484/2:getpid()= 11484 [11483] 11487: close(3)= 0 11484/2:write(10, 0x100147A64, 185) = 185 11484/2: T u e M a r 2 2 1 5 : 4 4 : 5 4 . 0 5 0 0 8 5 5 1 0 2 0 11484/2: 2 2 : p i d 1 1 4 8 4 : t h d - 1 0 0 1 5 5 0 0 0 : a m 11484/2: r e c o v e r : s e c u r i t y _ s e t e r r o r ( h a n d l 11484/2: e = 1 0 0 1 5 6 4 0 0 , d r i v e r = f f f f f f f f 7 1 3 9 11484/2: 2 f 4 0 ( B S D T C P ) e r r o r = s e n d m s g f a i l 11484/2: e d A : M e s s a g e t o o l o n g\n )\n -Ursprungligt meddelande- Från: Chris Hassell Skickat: den 21 mars 2022 20:37 Till: li...@xunil.at; ? amanda users ; Gunnarsson, Gunnar Ämne: RE: Sent out a pull-request to upgrade the packaging system NOTE: it seems at least gcc 7 is a good idea for the Solaris build. Autoconf 2.69 seems to be the minimal level. To start from a clean set of source. % git clean -dfx % ./autogen ( Not sure if autogen is necessary any more. It can be done automatically in the scripts. ) To build binaries and packages in the top directory (where the source is) ... depending what you are running on. [Same dir setup as old]. % ./packaging/rpm/build client % ./packaging/deb/build client % ./packaging/sun-pkg/build client To build current working dirs in a subdirectory
SV: Sent out a pull-request to upgrade the packaging system....
Hi, I had missed some packets but know autogen works. I'm getting this error. Thanks Gunnar libtool: link: gcc -Wall -Wextra -Wparentheses -Wdeclaration-after-statement -Wmissing-prototypes -Wstrict-prototypes -Wmissing-declarations -Wformat -Wformat-security -Wsign-compare -Wfloat-equal -Wold-style-definition -Wno-strict-aliasing -Wno-unknown-pragmas -Wno-deprecated-declarations "-DAMANDA_FILE=\"amflock-test.o\"" -g -O2 -fno-strict-aliasing -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -o .libs/amflock-test amflock-test.o -L/usr/lib/64 ./.libs/libamanda.so -L/usr/lib/sparcv9 ./.libs/libtestutils.a -lgobject-2.0 -lgthread-2.0 -lglib-2.0 -lsocket -lnsl -lresolv -lintl -lssl -lcrypto -R/opt/lib/amanda ld: fatal: symbol 'tu_debugging_enabled' is multiply-defined: (file amflock-test.o type=OBJT; file ./.libs/libtestutils.a(testutils.o) type=OBJT); collect2: error: ld returned 1 exit status gmake[3]: *** [Makefile:2371: amflock-test] Error 1 gmake[3]: Leaving directory '/root/amanda_compile/amanda-3_5-packaging-upgrade-wip/amanda/common-src' gmake[2]: *** [Makefile:2171: all] Error 2 gmake[2]: Leaving directory '/root/amanda_compile/amanda-3_5-packaging-upgrade-wip/amanda/common-src' gmake[1]: *** [Makefile:1737: all-recursive] Error 1 gmake[1]: Leaving directory '/root/amanda_compile/amanda-3_5-packaging-upgrade-wip/amanda' gmake: *** [Makefile:1661: all] Error 2 -Ursprungligt meddelande- Från: owner-amanda-us...@amanda.org För Gunnarsson, Gunnar Skickat: den 19 mars 2022 11:58 Till: li...@xunil.at; Chris Hassell ; ? amanda users Ämne: SV: Sent out a pull-request to upgrade the packaging system Hi, git clone -b 3_5-packaging-upgrade-wip https://github.com/zmanda/amanda.git git branch * 3_5-packaging-upgrade-wip On solaris these packages are needed pkg install git automake autoconf pkg:/developer/build/libtool gcc cd Amanda ./autogen + set -x + /usr/bin/autoconf configure.ac:25: error: possibly undefined macro: AC_DEFINE If this token and others are legitimate, please use m4_pattern_allow. See the Autoconf documentation. + die 'autoconf [default autoconf version] failed' + echo 'autoconf [default autoconf version] failed' autoconf [default autoconf version] failed + exit 1 This works on the master branch ? Thanks Gunnar -Ursprungligt meddelande- Från: owner-amanda-us...@amanda.org För Stefan G. Weichinger Skickat: den 18 mars 2022 10:31 Till: Chris Hassell ; ? amanda users Ämne: Re: Sent out a pull-request to upgrade the packaging system Am 17.03.22 um 22:28 schrieb Chris Hassell: > First of many. It is not configured-and-tested as is but it is very > close to our working version. > > -Versioning based on tags from git > > -An in-directory build as well as below-directory package build > > -Solaris 10 and Solaris 11 packages > > -Better handling of pre/post scripts (may need to update Debian > ones??) > > -Systemd services are included as well. > > If I can get someone to look over it and give it a spin I’d really > appreciate it. > > I have more fixes for the installcheck directory to allow some good > and solid self-tests. Nice to see progress. I assume you want us to try to build the code in that PR? pls point to some howto, I can't remember the procedure anymore (autoconf? etc) I can test on Debian 11, for example.
SV: Sent out a pull-request to upgrade the packaging system....
Hi, git clone -b 3_5-packaging-upgrade-wip https://github.com/zmanda/amanda.git git branch * 3_5-packaging-upgrade-wip On solaris these packages are needed pkg install git automake autoconf pkg:/developer/build/libtool gcc cd Amanda ./autogen + set -x + /usr/bin/autoconf configure.ac:25: error: possibly undefined macro: AC_DEFINE If this token and others are legitimate, please use m4_pattern_allow. See the Autoconf documentation. + die 'autoconf [default autoconf version] failed' + echo 'autoconf [default autoconf version] failed' autoconf [default autoconf version] failed + exit 1 This works on the master branch ? Thanks Gunnar -Ursprungligt meddelande- Från: owner-amanda-us...@amanda.org För Stefan G. Weichinger Skickat: den 18 mars 2022 10:31 Till: Chris Hassell ; ? amanda users Ämne: Re: Sent out a pull-request to upgrade the packaging system Am 17.03.22 um 22:28 schrieb Chris Hassell: > First of many. It is not configured-and-tested as is but it is very > close to our working version. > > -Versioning based on tags from git > > -An in-directory build as well as below-directory package build > > -Solaris 10 and Solaris 11 packages > > -Better handling of pre/post scripts (may need to update Debian > ones??) > > -Systemd services are included as well. > > If I can get someone to look over it and give it a spin I’d really > appreciate it. > > I have more fixes for the installcheck directory to allow some good > and solid self-tests. Nice to see progress. I assume you want us to try to build the code in that PR? pls point to some howto, I can't remember the procedure anymore (autoconf? etc) I can test on Debian 11, for example.
Amanda version 3.5.1 for 64B sparc 11.4
Hi, I'm testing out amanda version 3.5.1 for Sparc Solaris 11.4 ( native perl is know 64 ) and it compiles when using --without-ndmp ( no 64 support in Solaris ). I'm getting this error from amrecover. Any suggestion what might be the problem ? Should I compile the code with --without-ipv6 support ? /opt/sbin/amrecover AMRECOVER Version 3.5.1. Contacting server on XXX ... [request failed: sendmsg failed A: Message too long Thanks Gunnar Wed Mar 16 21:19:58.086282690 2022: pid 6672: thd-10017e450: amrecover: make_socket opening socket with family 2 Wed Mar 16 21:19:58.086309670 2022: pid 6672: thd-10017e450: amrecover: make_socket opening socket with family 2: 8 Wed Mar 16 21:19:58.087175210 2022: pid 6672: thd-10017e450: amrecover: ambind failed: sendmsg failed A: Message too long Wed Mar 16 21:19:58.087238600 2022: pid 6672: thd-10017e450: amrecover: stream_client: Could not bind to port in range 512-1023. Wed Mar 16 21:19:58.087252660 2022: pid 6672: thd-10017e450: amrecover: security_seterror(handle=10017da20, driver=77b93b60 (B SDTCP) error=sendmsg failed A: Message too long
SV: Solaris 11.4 64 binaries
Hi Chris thanks for the update. Let me know if you need help testing out the Solaris 11 port with the 64 binaries. Thanks Gunnar Från: Chris Hassell Skickat: den 10 november 2021 05:36 Till: Gunnarsson, Gunnar ; amanda-users@amanda.org Ämne: RE: Solaris 11.4 64 binaries I have. We have a community release due… way … way way overdue. I’m afraid it’s been only myself doing it.(A lot of change has been disruptive, I’m afraid). In addition... I got the packaging to work for the new Solaris 11 type. So there’s that. When we have version 4.1.1 out I’ll have more time. n Christopher Hassell, Betsol Corporation From: owner-amanda-us...@amanda.org<mailto:owner-amanda-us...@amanda.org> mailto:owner-amanda-us...@amanda.org>> On Behalf Of Gunnarsson, Gunnar Sent: Tuesday, November 2, 2021 1:40 AM To: amanda-users@amanda.org<mailto:amanda-users@amanda.org> Subject: Solaris 11.4 64 binaries WARNING: This email originated from outside of BETSOL. Do not click links or open attachments unless you recognize the sender and know the content is safe. Hi I’m been trying to compile the latest community release for Solaris for 64 binaries, Solaris 11.4 ships with Perl 64 binaries. Has anyone done this ? Are there any more community releases due ? Thanks Gunnar Gunnarsson Skickat från min iPhone Confidentiality Notice | The information transmitted by this email is intended only for the person or entity to which it is addressed. This email may contain proprietary, business-confidential and/or privileged material. If you are not the intended recipient of this message, be aware that any use, review, re-transmission, distribution, reproduction or any action taken in reliance upon this message is strictly prohibited. If you received this in error, please contact the sender and delete the material from all computers.
Solaris 11.4 64 binaries
Hi I’m been trying to compile the latest community release for Solaris for 64 binaries, Solaris 11.4 ships with Perl 64 binaries. Has anyone done this ? Are there any more community releases due ? Thanks Gunnar Gunnarsson Skickat från min iPhone
amvault - using vault-storage section
Hello amanda-users,
When I upgraded to amanda version 3.4.5 I added a storage section in order to
run the amvault and it works. But the daily runs are complaining about not
finding any volumes in that storage sections as they are marked no-reuse. So
what is wrong with my setup ?
when I'm getting the error below after each daily runs.
These dumps were to tapes Orabck-Unix-006, Orabck-Unix-007.
*** A TAPE ERROR OCCURRED: [No acceptable volumes found].
Some dumps may have been left in the holding disk.
The next 3 tapes Amanda expects to use are: Orabck-Unix-008, Orabck-Unix-009,
Orabck-Unix-010.
vault-storage "vault-st"
#
# define storage
define storage "vault-st" {
autolabel "Orabck-Unix-2%%" empty
tapetype "HP-Ultrium3"
tpchanger "vault"
}
Thanks Gunnar Gunnarsson
amtapetype - amtapetype: Error setting FSF_AFTER_FILEMARK
Hello amanda-users, I've been testing the the latest version 3.4.5 version and got som issues with amtapetype. I'm running on Solaris 11 Sparc # /opt/sbin/amtapetype -f /dev/rmt/2n Checking for FSF_AFTER_FILEMARK requirement amtapetype: Error setting FSF_AFTER_FILEMARK: Not allowed to set property at /opt/sbin/amtapetype line 430. In the changer we define device_property "FSF_AFTER_FILEMARK" "TRUE" do I have to specify it ? # mt -f /dev/rmt/2n status HP Ultrium LTO 4 tape drive: sense key(0x0)= No Additional Sense residual= 0 retries= 0 file no= 1 block no= 0 Thanks Gunnar Gunnarsson
Build problem amanda-3.4.2 sparc Solaris 11
Hi I'm building amanda-3.4.2 on SunOS 5.11 11.2 sun4v sparc sun4v using gcc: gcc --version gcc (GCC) 4.5.2 cal/lib/amanda Undefined first referenced symbol in file get_sse42 ./.libs/libamanda.so ld: fatal: symbol referencing errors collect2: ld returned 1 exit status gmake[1]: *** [amflock-test] Error 1 gmake[1]: Leaving directory `/var/tmp/amanda-3.4.2/common-src' gmake: *** [all] Error 2 Thanks Gunnar Gunnarsson
SV: Tape write rate much slower efter upgrade
Hi, Below tape performance from running similar workload in version 3.2.0 and recently upgraded version 3.3.7p1. In this run I'm doing tape to tape vaulting. As you can see a huge degraded writing performance. I'm running the backup on Solaris 10 on sparc. Xfer IO seems to be much slower than in previous version for some reason ? Any ideas how to proceed and find out what's going on ? Thanks Gunnar Gunnarsson 3.2.0 Tape Time (hrs:min) 3:02 3:02 0:00 Tape Size (meg) 747221.3 747221.30.0 Tape Used (%) 170.2 170.20.0 DLEs Taped16 16 0 Parts Taped 29 29 0 Avg Tp Write Rate (k/s) 70005.070005.0-- USAGE BY TAPE: Label Time Size % DLEs Parts Orabck-Unix-194 1:49 440059M 94.41218 Orabck-Unix-195 1:11 307162M 65.9 411 3.3.7p1 Tape Time (hrs:min) 4:54 4:54 0:00 Tape Size (meg) 598327.5 598327.50.0 Tape Used (%) 135.9 135.90.0 DLEs Taped11 11 0 Parts Taped 23 23 0 Avg Tp Write Rate (k/s) 34738.834738.8-- USAGE BY TAPE: Label Time Size % DLEs Parts Orabck-Unix-196 3:19 410261M 88.0 715 Orabck-Unix-197 1:32 188066M 40.4 4 8 -Ursprungligt meddelande- Från: owner-amanda-us...@amanda.org [mailto:owner-amanda-us...@amanda.org] För Gunnarsson, Gunnar Skickat: den 3 september 2015 22:11 Till: Jean-Francois Malouin; amanda-users@amanda.org Ämne: SV: Tape write rate much slower efter upgrade I'm doing amvault from disk to tape the workload on the server is the same no shortage of memory cpu resources. Xfer rate to tape reduced by 60% from version 3.2.0 to 3.3.7 ! -Ursprungligt meddelande- Från: owner-amanda-us...@amanda.org [mailto:owner-amanda-us...@amanda.org] För Jean-Francois Malouin Skickat: den 3 september 2015 21:37 Till: amanda-users@amanda.org Ämne: Re: Tape write rate much slower efter upgrade Hi, * Jean-Louis Martineau <martin...@zmanda.com> [20150903 15:21]: > Gunnar, > > Do the number by DLE always differ? > Are you using a holding disk? > Can you monitor the server to find some bottleneck, memory, cpu, > network, disk I/O, ... I witnessed the same behaviour back in February when upgrading from 3.3.5 to 3.3.7. I tried to post to the list a few times but somehow my posts never made it, maybe because they had 2 pics attached showing the performance from the hold disk to the tape drive. Here's what I wrote back then: "2 weeks ago I upgraded one of my amanda server from 3.3.5 to 3.3.7. Right away I noticed a ~25% tape write performance degradation to a LTO4 tape drive. I've attached 2 pics showing the xfer IO rate to/from the holddisk: one shows a snapshot of the week after the upgrade with the 10th of Feb still at 3.3.5 (and a xfer rate of ~100MBs) and a drop to ~75MBs after the upgrade was done. I reverted back to 3.3.5 a few days ago and the 2nd snapshot shows a return to previous rate." HTH, jf > > Jean-Louis > > On 03/09/15 02:49 PM, Gunnarsson, Gunnar wrote: > > > >Hi, > > > >See below slow down in avg tp write rate after having upgraded amanda > >from 3.2.0 to 3.3.7p1 nothing else is changed. > > > >We are using DLT drivers SCSI and fiber connected. > > > >Thanks Gunnar > > > >Tape Time (hrs:min) 2:37 2:37 0:00 > > > >Tape Size (meg) 382968.9 382968.90.0 > > > >Tape Used (%) 324.1 324.10.0 > > > >DLEs Taped 462462 0 > > > >Parts Taped 462462 0 > > > >Avg Tp Write Rate (k/s) 41670.441670.4-- > > > >(brought to you by Amanda version 3.2.0) > > > >Tape Time (hrs:min) 4:18 4:18 0:00 > > > >Tape Size (meg) 384525.8 384525.80.0 > > > >Tape Used (%) 334.0 334.00.0 > > > >DLEs Taped 481481 0 > > > >Parts Taped 481481 0 > > > >Avg Tp Write Rate (k/s) 25469.225469.2-- > > > >(brought to you by Amanda version 3.3.7p1) > > > >Tape Time (hrs:min) 6:03 1:46 4:18 > > > >Tape Size (meg) 710405.4 208727.0 501678.5 > > > >Tape Used (%) 161.0 47.4 113.6 > > > >DLEs Taped 12 4 8 1:7 2:1 > > > >Parts Taped 27 8 19 1:11 2:8 > > > >Avg Tp Write Rate (k/s) 33357.333701.733216.0 > > > >USAGE BY TAPE: > > > > Label Time Size % DLEs Part
Tape write rate much slower efter upgrade
Hi, See below slow down in avg tp write rate after having upgraded amanda from 3.2.0 to 3.3.7p1 nothing else is changed. We are using DLT drivers SCSI and fiber connected. Thanks Gunnar Tape Time (hrs:min) 2:37 2:37 0:00 Tape Size (meg) 382968.9 382968.90.0 Tape Used (%) 324.1 324.10.0 DLEs Taped 462462 0 Parts Taped 462462 0 Avg Tp Write Rate (k/s) 41670.441670.4-- (brought to you by Amanda version 3.2.0) Tape Time (hrs:min) 4:18 4:18 0:00 Tape Size (meg) 384525.8 384525.80.0 Tape Used (%) 334.0 334.00.0 DLEs Taped 481481 0 Parts Taped 481481 0 Avg Tp Write Rate (k/s) 25469.225469.2-- (brought to you by Amanda version 3.3.7p1) Tape Time (hrs:min) 6:03 1:46 4:18 Tape Size (meg) 710405.4 208727.0 501678.5 Tape Used (%) 161.0 47.4 113.6 DLEs Taped12 4 8 1:7 2:1 Parts Taped 27 8 19 1:11 2:8 Avg Tp Write Rate (k/s) 33357.333701.733216.0 USAGE BY TAPE: Label Time Size % DLEs Parts Orabck-Unix-002 3:37 431135M 92.51119 Orabck-Unix-003 2:25 279270M 59.9 1 8 (brought to you by Amanda version 3.3.7p1) Tape Time (hrs:min) 3:13 0:11 3:01 Tape Size (meg) 776730.950668.0 726062.8 Tape Used (%) 177.9 12.6 165.3 DLEs Taped18 4 14 1:9 2:5 Parts Taped 33 4 29 1:11 2:18 Avg Tp Write Rate (k/s) 68857.477438.968329.0 USAGE BY TAPE: Label Time Size % DLEs Parts Orabck-Unix-015 1:34 392152M 84.2 412 Orabck-Unix-016 1:34 384579M 82.51421 (brought to you by Amanda version 3.2.0)
SV: Tape write rate much slower efter upgrade
I'm doing amvault from disk to tape the workload on the server is the same no shortage of memory cpu resources. Xfer rate to tape reduced by 60% from version 3.2.0 to 3.3.7 ! -Ursprungligt meddelande- Från: owner-amanda-us...@amanda.org [mailto:owner-amanda-us...@amanda.org] För Jean-Francois Malouin Skickat: den 3 september 2015 21:37 Till: amanda-users@amanda.org Ämne: Re: Tape write rate much slower efter upgrade Hi, * Jean-Louis Martineau <martin...@zmanda.com> [20150903 15:21]: > Gunnar, > > Do the number by DLE always differ? > Are you using a holding disk? > Can you monitor the server to find some bottleneck, memory, cpu, > network, disk I/O, ... I witnessed the same behaviour back in February when upgrading from 3.3.5 to 3.3.7. I tried to post to the list a few times but somehow my posts never made it, maybe because they had 2 pics attached showing the performance from the hold disk to the tape drive. Here's what I wrote back then: "2 weeks ago I upgraded one of my amanda server from 3.3.5 to 3.3.7. Right away I noticed a ~25% tape write performance degradation to a LTO4 tape drive. I've attached 2 pics showing the xfer IO rate to/from the holddisk: one shows a snapshot of the week after the upgrade with the 10th of Feb still at 3.3.5 (and a xfer rate of ~100MBs) and a drop to ~75MBs after the upgrade was done. I reverted back to 3.3.5 a few days ago and the 2nd snapshot shows a return to previous rate." HTH, jf > > Jean-Louis > > On 03/09/15 02:49 PM, Gunnarsson, Gunnar wrote: > > > >Hi, > > > >See below slow down in avg tp write rate after having upgraded amanda > >from 3.2.0 to 3.3.7p1 nothing else is changed. > > > >We are using DLT drivers SCSI and fiber connected. > > > >Thanks Gunnar > > > >Tape Time (hrs:min) 2:37 2:37 0:00 > > > >Tape Size (meg) 382968.9 382968.90.0 > > > >Tape Used (%) 324.1 324.10.0 > > > >DLEs Taped 462462 0 > > > >Parts Taped 462462 0 > > > >Avg Tp Write Rate (k/s) 41670.441670.4-- > > > >(brought to you by Amanda version 3.2.0) > > > >Tape Time (hrs:min) 4:18 4:18 0:00 > > > >Tape Size (meg) 384525.8 384525.80.0 > > > >Tape Used (%) 334.0 334.00.0 > > > >DLEs Taped 481481 0 > > > >Parts Taped 481481 0 > > > >Avg Tp Write Rate (k/s) 25469.225469.2-- > > > >(brought to you by Amanda version 3.3.7p1) > > > >Tape Time (hrs:min) 6:03 1:46 4:18 > > > >Tape Size (meg) 710405.4 208727.0 501678.5 > > > >Tape Used (%) 161.0 47.4 113.6 > > > >DLEs Taped 12 4 8 1:7 2:1 > > > >Parts Taped 27 8 19 1:11 2:8 > > > >Avg Tp Write Rate (k/s) 33357.333701.733216.0 > > > >USAGE BY TAPE: > > > > Label Time Size % DLEs Parts > > > > Orabck-Unix-002 3:37 431135M 92.51119 > > > > Orabck-Unix-003 2:25 279270M 59.9 1 8 > > > >(brought to you by Amanda version 3.3.7p1) > > > >Tape Time (hrs:min) 3:13 0:11 3:01 > > > >Tape Size (meg) 776730.950668.0 726062.8 > > > >Tape Used (%) 177.9 12.6 165.3 > > > >DLEs Taped 18 4 14 1:9 2:5 > > > >Parts Taped 33 4 29 1:11 2:18 > > > >Avg Tp Write Rate (k/s) 68857.477438.968329.0 > > > >USAGE BY TAPE: > > > > Label Time Size % DLEs Parts > > > > Orabck-Unix-015 1:34 392152M 84.2 412 > > > > Orabck-Unix-016 1:34 384579M 82.51421 > > > >(brought to you by Amanda version 3.2.0) > > >
amvault failing
Hi, I've been running amvault successfully for serveral years with version amvault-3.2.3. Today I got the fatal in ARRAY see below. Only changes made to the system recently is bash patch. I'm running Solaris 5.10 sparc. amvault --dry-run -o mailto=vault --fulls-only --src-timestamp 20140910-20140928 \ --dst-changer vault --label-template daily-1%% --autolabel empty daily [!^t]** bash-3.00$ cat amvault.20141003135147.debug Fri Oct 3 13:52:17 2014: amvault: warning: Use of uninitialized value in string comparison (cmp) at /usr/perl5/site_perl/5.8.4/Amanda/Recovery/Planner.pm line 298. Fri Oct 3 13:52:17 2014: amvault: warning: Use of uninitialized value in string comparison (cmp) at /usr/perl5/site_perl/5.8.4/Amanda/Recovery/Planner.pm line 298. Fri Oct 3 13:52:17 2014: amvault: warning: Use of uninitialized value in string comparison (cmp) at /usr/perl5/site_perl/5.8.4/Amanda/Recovery/Planner.pm line 298. Fri Oct 3 13:52:17 2014: amvault: warning: Use of uninitialized value in string comparison (cmp) at /usr/perl5/site_perl/5.8.4/Amanda/Recovery/Planner.pm line 298. Fri Oct 3 13:52:17 2014: amvault: warning: Use of uninitialized value in string comparison (cmp) at /usr/perl5/site_perl/5.8.4/Amanda/Recovery/Planner.pm line 298. Fri Oct 3 13:52:17 2014: amvault: warning: Use of uninitialized value in string comparison (cmp) at /usr/perl5/site_perl/5.8.4/Amanda/Recovery/Planner.pm line 298. Fri Oct 3 13:52:17 2014: amvault: warning: Use of uninitialized value in string comparison (cmp) at /usr/perl5/site_perl/5.8.4/Amanda/Recovery/Planner.pm line 298. Fri Oct 3 13:52:17 2014: amvault: critical (fatal): Can't use an undefined value as an ARRAY reference at /opt/sfw/sbin/amvault line 305. Thanks Gunnar
Support for LVM snapshot revisited ( amlvm-snapshot )
Hi,
I recently installed LVM snapshot application amlvm-snapshot from
https://github.com/marxarelli/amlvm-snapshot
Having fixed minor problem with the kernel device mapper paths I got it up and
running , substitute ../ with /dev and cleaning unused mount point.
I'm running Ubuntu 10.04.3 LTS with Ubuntu Amanda packages version 2.6.1p1.
Why not include this application in the core Amanda distribution properly ?
--GG
# /usr/lib/amanda/application# diff -c amlvm-snapshot amlvm-snapshot.orig
*** amlvm-snapshot 2012-01-16 08:34:20.138353314 +0100
--- amlvm-snapshot.orig 2012-01-13 14:20:27.188361522 +0100
***
*** 207,213
# executed with elevated privileges (sudo).
my $real_path = join(, $self-execute(1, readlink, $path));
chomp($real_path);
- $real_path =~ s@\.\.@/dev@mailto:s@\.\.@/dev@;
return ($real_path ne ) ? $real_path : $path;
}
--- 207,212
***
*** 348,355
sub umount_snapshot {
my $self = shift;
my $device = $self-readlink($self-get_snap_device());
- $device =~ s@\.\.@/dev@mailto:s@\.\.@/dev@;
- debug(umount_snapshot $device);
my $mnt = $self-scan_mtab(sub { return $_[1] if ($_[0] eq $device); });
--- 347,352
***
*** 361,369
}
$self-execute(1, umount, $mnt);
debug(Un-mounted snapshot device `$device' from `$mnt'.);
- rmdir $mnt;
- debug(Remove snapshot mount point rmdir `$mnt'.);
}
--- 358,365
}
$self-execute(1, umount, $mnt);
+
debug(Un-mounted snapshot device `$device' from `$mnt'.);
}
/etc/sudoers.d/amanda
%backup ALL=(ALL) NOPASSWD: /sbin/lvcreate, /sbin/lvdisplay, /sbin/lvremove,
/sbin/vgdisplay, /bin/readlink, /bin/mount, /bin/umount
Mount and umount was missing in the Readme file.
SV: amvault for disk-to-tape-to-tape setups?
Hi Sven, We are vaulting from disk to tape and from tape to tape and its working. We would like to see some more filtering option but apart from that its working quite well. Define a dst-changer definition in your amanda.conf and run the amvault command and try it out with the --dry-run option . amvault --dry-run -o mailto=vault \ --fulls-only --src-timestamp 20111224-20111230 \ --dst-changer vault --label-template Orabck-Unix-1%% \ --autolabel empty --GG -Ursprungligt meddelande- Från: owner-amanda-us...@amanda.org [mailto:owner-amanda-us...@amanda.org] För Sven Rudolph Skickat: den 13 januari 2012 13:43 Till: amanda-users@amanda.org Ämne: amvault for disk-to-tape-to-tape setups? Hello, every month I run full backups to tape, and I want to make an extra copy of these tapes. I guess that amvault could create these copies and make amanda aware of these copies. On the mailing list I mainly found references where amvault is used in disk-to-disk-to-tape setups, where the backup is done to disk and amvault creates tapes from the disk backup. Is amvault expected to create tapes from an existing tape backup? Is anyone doing this? Successfully? Thanks, Sve
SV: amvault question
I'm excluding all machines that start with the letter t or u, '[^tu]' but then I need to add back the host ux5042. How does the dot (.) host sperator work in amvault ? Thanks for any help Gunnar Gunnarsson $amvault --dry-run -o mailto=vault -o tapetype=HP-Ultrium3 --fulls-only --src-timestamp 20110711-31 \ --dst-changer vault --label-template daily-1%% --autolabel any \ daily '[!^tu]*' snipped daily-2 55 delfi2 /cots 20110731200102 0 5451280 Total Size: 936995691 KB
amvault question
Hi, I'm runing amvault excluding host starting with t or u but I want to keep one u machine as well. How can I specify this with amvault ? I though that this might work out with this but apparently not. $ amvault --dry-run -o mailto=vault -o tapetype=HP-Ultrium3 --fulls-only --src-timestamp 20110711-20110731 \ --dst-changer vault --label-template daily-1%% --autolabel any daily '[!^tu]*' ux5042 '*' Total Size: 0 KB $amvault --dry-run -o mailto=vault -o tapetype=HP-Ultrium3 --fulls-only --src-timestamp 20110711-31 \ --dst-changer vault --label-template daily-1%% --autolabel any \ daily '[!^tu]*' snipped daily-2 55 delfi2 /cots 20110731200102 0 5451280 Total Size: 936995691 KB -- Thanks for any help Gunnar Gunnarsson
amtapetype question
I'm running Solaris sparc 10 with $ mt -f /dev/rmt/0n usage: mt [ -f device ] command [ count ] bash-3.00$ mt -f /dev/rmt/0n status HP Ultrium LTO 3 tape drive: sense key(0x0)= No Additional Sense residual= 0 retries= 0 file no= 1 block no= 0 $ amtapetype -f tape:/dev/rmt/0n Checking for FSF_AFTER_FILEMARK requirement amtapetype: Error setting FSF_AFTER_FILEMARK: Success at /opt/sfw/sbin/amtapetype line 419. What does this mean isn't this set default in Solaris ? -- Thanks Gunnar Gunnarsson
LEOM and part_size question
Hi, If LEOM is working on my storage device should I define part_size as well ? Doing disk backup ( Vtapes ) LEOM is automatically enabled and I don't specify any part_size. What about tape storage device should I add a part_size ? -- Thanks Gunnar Gunnarsson
amvault-3.2.3
Hi, Upgraded to 3.2.3 from 3.2.0 to fix split and LEOM handling. amvault is not listning the size of the dumps see below. I'm running Solaris sparc 10 perl version v5.8.4. amvault --dry-run -o mailto=vault --fulls-only --src-timestamp 20110709-20110801 --dst-changer vault --label-template daily-1%% --autolabel any daily tx5019 daily-19 94 tx5019 / 20110710200103 0 daily-24 29 tx5019 /export/home 20110714200102 0 daily-38 114 tx5019 /export 20110723200103 0 daily-41 161 tx5019 / 20110725200103 0 daily-45 26 tx5019 /export/home 20110729200103 0 Total Size: NaN KB Thanks Gunnar Gunnarsson
SV: amvault-3.2.3
ok type casting to int solved my problem with Total size printing.
$total_kb += int $dump-{'kb'};
---
$total_kb += $dump-{'kb'};
Thanks Gunar Gunnarsson
Från: owner-amanda-us...@amanda.org [mailto:owner-amanda-us...@amanda.org] För
Gunnarsson, Gunnar
Skickat: den 2 augusti 2011 11:43
Till: amanda-users@amanda.org
Ämne: amvault-3.2.3
Hi,
Upgraded to 3.2.3 from 3.2.0 to fix split and LEOM handling.
amvault is not listning the size of the dumps see below. I'm running Solaris
sparc 10 perl version v5.8.4.
amvault --dry-run -o mailto=vault --fulls-only --src-timestamp
20110709-20110801 --dst-changer vault --label-template daily-1%% --autolabel
any daily tx5019
daily-19 94 tx5019 / 20110710200103 0
daily-24 29 tx5019 /export/home 20110714200102 0
daily-38 114 tx5019 /export 20110723200103 0
daily-41 161 tx5019 / 20110725200103 0
daily-45 26 tx5019 /export/home 20110729200103 0
Total Size: NaN KB
Thanks Gunnar Gunnarsson
SV: amvault $config_overrides
Hi,
I tested the patch and got some side effects in amreport see below. As you can
see the date is incorrect.
Thanks Gunnar Gunnarsson
Tue Nov 16 12:24:29 2010: amreport: pid 22412 ruid 91 euid 91 version 3.2.0:
start at Tue Nov 16 12:24:29 2010
Tue Nov 16 12:24:29 2010: amreport: config_overrides: mailto vault
Tue Nov 16 12:24:29 2010: amreport: pid 22412 ruid 91 euid 91 version 3.2.0: ren
ame at Tue Nov 16 12:24:29 2010
Tue Nov 16 12:24:29 2010: amreport: using logfile: /var/opt/amanda/orabck/log
Tue Nov 16 12:24:29 2010: amreport: operating in script mode
Tue Nov 16 12:24:29 2010: amreport: detected an amvault run
Tue Nov 16 12:24:29 2010: amreport: planned output: human mail vault
Tue Nov 16 12:24:29 2010: amreport: warning: Use of uninitialized value in
numeric gt () at /opt/sfw/sbin/amreport line 368.
Tue Nov 16 12:24:29 2010: amreport: warning: Use of uninitialized value in int
at /opt/sfw/sbin/amreport line 369.
Tue Nov 16 12:24:29 2010: amreport: invoking mail app: /usr/ucb/Mail -s SVK
Unix AMVAULT MAIL REPORT FOR Dec 2, vault
Tue Nov 16 12:24:33 2010: amreport: warning: Use of uninitialized value in
numeric lt () at /usr/perl5/site_perl/5.8.4/Amanda/Report/human.pm line 1282.
Try the attached patch.
Jean-Louis
Gunnarsson, Gunnar wrote:
Hi,
I a minor issue regarding $config_overrides that are not beeing
propagated to the call in amreport thus ignoring setting such as
tapettype mailto etc.
debug(invoking amreport..);
system($sbindir/amreport, $self-{'config_name'}, --from-amdump\
);
amvault $config_overrides
Hi,
I a minor issue regarding $config_overrides that are not beeing propagated to
the call in amreport thus ignoring setting
such as tapettype mailto etc.
debug(invoking amreport..);
system($sbindir/amreport, $self-{'config_name'}, --from-amdump\
);
Thanks Gunnar Gunnarsson
SV: amgtar not using directory PROPERTY
Jean-Louis, Thanks for attached patch - what a difference a line of code can do. Now it's working. Thanks Gunnar Gunnarsson Gunnar, Try the attached patch. Jean-Louis
amgtar not using directory PROPERTY
Hi,
I version 3.2.0 amgtar is not using PROPERTY directory amzfs-snapshot, has
something changed regarding this in 3.2.0 ?
This has been working in the previous version of amanda, snipped from debug
file:
sendbackup: script: PROPERTY directory
/export/.zfs/snapshot/amanda-_export-current
amgtar: Spawning /usr/local/bin/tar /usr/local/bin/tar --create --verbose
--file - --directory /export --one-file-system --no-check-device
--listed-incremental /var/opt/lib/amanda/gnutar-lists/tx5019_export_1.new
--sparse --ignore-failed-read --totals . in pipeline
amgtar should be called with --directory
/export/.zfs/snapshot/amanda-_export-current
define dumptype snap-zfs-amgtar {
dt_amgtar
script amzfs_snapshot
maxdumps 2
}
Thanks Gunnar Gunnarsson
Amanda 3.2.0 zfs Solaris client
HI, I'm testing the 3.2.0 client on Solaris 10, using Amanda server 3.1.1 using zfs with dumptype using amzfs_snapshot and gettgin: tx5019 / lev 1 FAILED [/usr/local/bin/tar exited with status 2: see /tmp/amanda/client/daily/amgtar.20101023202125000.debug] Snapshot are beeing created but not used see below in the history list and the debug file. Is the client incorrectly installed or is this a bug ? Thanks Gunnar Gunnarsson bash-3.00# zpool history | grep 10-23 2010-10-23.20:01:19 zfs snapshot rpool/ROOT/zfs1...@amanda-_-current 2010-10-23.20:03:46 zfs destroy rpool/ROOT/zfs1...@amanda-_-current 2010-10-23.20:03:48 zfs snapshot rpool/exp...@amanda-_export-current 2010-10-23.20:03:49 zfs destroy rpool/exp...@amanda-_export-current 2010-10-23.20:03:50 zfs snapshot rpool/export/h...@amanda-_export_home-current 2010-10-23.20:04:01 zfs destroy rpool/export/h...@amanda-_export_home-current 2010-10-23.20:12:07 zfs snapshot rpool/export/h...@amanda-_export_home-current 2010-10-23.20:12:55 zfs destroy rpool/export/h...@amanda-_export_home-current 2010-10-23.20:12:56 zfs snapshot rpool/ROOT/zfs1...@amanda-_-current 2010-10-23.20:21:21 zfs destroy rpool/ROOT/zfs1...@amanda-_-current 2010-10-23.20:21:24 zfs snapshot rpool/ROOT/zfs1...@amanda-_-current 2010-10-23.20:29:46 zfs destroy rpool/ROOT/zfs1...@amanda-_-current 2010-10-23.21:03:16 zfs snapshot rpool/exp...@amanda-_export-current 2010-10-23.21:03:17 zfs destroy rpool/exp...@amanda-_export-current This setup is working with the 3.1.1 client. /tmp/amanda/client/daily/amgtar.20101023202125000.debug: Sat Oct 23 20:21:25 2010: amgtar: pid 387 ruid 0 euid 90 version 3.2.0: start at Sat Oct 23 20:21:25 2010 Sat Oct 23 20:21:25 2010: amgtar: version 3.2.0 Sat Oct 23 20:21:25 2010: amgtar: pid 387 ruid 0 euid 90 version 3.2.0: rename at Sat Oct 23 20:21:25 2010 Sat Oct 23 20:21:25 2010: amgtar: GNUTAR-PATH /usr/local/bin/tar Sat Oct 23 20:21:25 2010: amgtar: GNUTAR-LISTDIR /var/opt/lib/amanda/gnutar-lists Sat Oct 23 20:21:25 2010: amgtar: ONE-FILE-SYSTEM yes Sat Oct 23 20:21:25 2010: amgtar: SPARSE yes Sat Oct 23 20:21:25 2010: amgtar: NO-UNQUOTE no Sat Oct 23 20:21:25 2010: amgtar: ATIME-PRESERVE no Sat Oct 23 20:21:25 2010: amgtar: ACLS no Sat Oct 23 20:21:25 2010: amgtar: SELINUX no Sat Oct 23 20:21:25 2010: amgtar: XATTRS no Sat Oct 23 20:21:25 2010: amgtar: CHECK-DEVICE no Sat Oct 23 20:21:25 2010: amgtar: SIZE ^ *Total bytes written: [0-9][0-9]* Sat Oct 23 20:21:25 2010: amgtar: IGNORE : Directory is new$ Sat Oct 23 20:21:25 2010: amgtar: IGNORE : Directory has been renamed Sat Oct 23 20:21:25 2010: amgtar: IGNORE file changed as we read it Sat Oct 23 20:21:25 2010: amgtar: IGNORE Cannot stat: No such file or directory Sat Oct 23 20:21:25 2010: amgtar: IGNORE Warning: Cannot stat: No such file or directory Sat Oct 23 20:21:25 2010: amgtar: NORMAL ^could not open conf file Sat Oct 23 20:21:25 2010: amgtar: NORMAL ^Elapsed time: Sat Oct 23 20:21:25 2010: amgtar: NORMAL ^Throughput Sat Oct 23 20:21:25 2010: amgtar: NORMAL : socket ignored$ Sat Oct 23 20:21:25 2010: amgtar: NORMAL : File .* shrunk by [0-9][0-9]* bytes, padding with zeros Sat Oct 23 20:21:25 2010: amgtar: NORMAL : Cannot add file .*: No such file or directory$ Sat Oct 23 20:21:25 2010: amgtar: NORMAL : Error exit delayed from previous errors Sat Oct 23 20:23:23 2010: amgtar: Spawning /usr/local/bin/tar /usr/local/bin/tar --create --verbose --file - --directory / --one-file-system --no-check-device listed-incremental /var/opt/lib/amanda/gnutar-lists/tx5019__1.new--sparse --ignore-failed-read --totals . in pipeline Sat Oct 23 20:23:30 2010: amgtar: 0: strange(?): /usr/local/bin/tar: ./system/contract/process/111: Cannot savedr: No such file or directory Sat Oct 23 20:23:30 2010: amgtar: 0: strange(?): /usr/local/bin/tar: ./system/contract/process/113: Cannot savedir: No such file or directory etc.
SV: amvault - version 3.2.0beta3
Yes, I think so. Did you sort this out? Thanks Gunnar Gunnarsson
SV: amvault report - *** THE DUMPS DID NOT FINISH PROPERLY!
This has worked in all previous version of Amanda. . Ah, so this is using the amzfs-snapshot scripts? Jean-Louis, did something change here? I don't recall if we have a way for a script to change the diskdevice for a DLE? Did that work in 3.1 and not in 3.2? Tnanks Gunnar Gunnarson
SV: amvault report - *** THE DUMPS DID NOT FINISH PROPERLY!
OK. You should adjust the length in your tapetype definition, then - your
tape appears to be at least 550% longer than you have configured.
Can I add the tape length to:
define changer vault {
tpchanger chg-manual
tapedev tape:/dev/rmt/0n
changerfile changer.conf
}
Not sure what you mean by this - what was delayed?
The holding disk was filled up before vtape was written.
You probably should be excluding ./devices?
This worked in verions 3.1.1 something is changed 3.2.0beta3 nothing else has
changed on the system. I testaed another Solaris client with this version and
got the same problem.
FAILED DUMP DETAILS:
/-- hansabck / lev 1 FAILED [/usr/local/bin/tar exited with status 2: see
/tmp/amanda/client/hansa/amgtar.20101008223221000.debug]
sendbackup: info BACKUP=APPLICATION
sendbackup: info APPLICATION=amgtar
sendbackup: info RECOVER_CMD=/usr/bin/gzip -dc
|/opt/sfw/libexec/amanda/application/amgtar restore [./file-to-restore]+
sendbackup: info COMPRESS_SUFFIX=.gz
sendbackup: info end
? /usr/local/bin/tar: ./devices/ramdisk-root: Cannot savedir: No such file or
directory
? /usr/local/bin/tar: ./devices/ramdisk-root/: Warning: Cannot stat: No such
file or directory
? /usr/local/bin/tar: ./dev/fd: file changed as we read it
? /usr/local/bin/tar: ./devices: file changed as we read it
? /usr/local/bin/tar: ./devices/e...@c0: file changed as we read it
? /usr/local/bin/tar: ./devices/e...@c0/ser...@0,ca: file changed as we
read it
? /usr/local/bin/tar: ./devices/iscsi: file changed as we read it
? /usr/local/bin/tar: ./devices/n...@80: file changed as we read it
? /usr/local/bin/tar: ./devices/options: file changed as we read it
? /usr/local/bin/tar: ./devices/pci-performance-count...@0: file changed as
we read it
? /usr/local/bin/tar: ./devices/p...@0: file changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0: file changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@1: file changed as we read
it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@1/p...@0: file changed as
we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@1/p...@0/p...@1: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@1/p...@0/p...@1/p...@0:
file changed as we read it
? /usr/local/bin/tar:
./devices/p...@0/p...@0/p...@1/p...@0/p...@1/p...@0/u...@0: file changed as we
read it
? /usr/local/bin/tar:
./devices/p...@0/p...@0/p...@1/p...@0/p...@1/p...@0/u...@0,1: file changed as
we read it
? /usr/local/bin/tar:
./devices/p...@0/p...@0/p...@1/p...@0/p...@1/p...@0/u...@0,2: file changed as
we read it
? /usr/local/bin/tar:
./devices/p...@0/p...@0/p...@1/p...@0/p...@1/p...@0/u...@0,2/h...@4: file
changed as we read it
? /usr/local/bin/tar:
./devices/p...@0/p...@0/p...@1/p...@0/p...@1/p...@0/u...@0,2/stor...@2: file
changed as we read it
? /usr/local/bin/tar:
./devices/p...@0/p...@0/p...@1/p...@0/p...@1/p...@0/u...@0,2/stor...@2/d...@0,0:
file changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@1/p...@0/p...@2: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@1/p...@0/p...@2/netw...@0:
file changed as we read it
? /usr/local/bin/tar:
./devices/p...@0/p...@0/p...@1/p...@0/p...@2/netw...@0,1: file changed as we
read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@1/p...@0/p...@3: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@1/p...@0/p...@3/netw...@0:
file changed as we read it
? /usr/local/bin/tar:
./devices/p...@0/p...@0/p...@1/p...@0/p...@3/netw...@0,1: file changed as we
read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@2: file changed as we read
it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@2/s...@0: file changed as
we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@2/s...@0/s...@0,0: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@2/s...@0/s...@1,0: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@2/s...@0/s...@2,0: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@2/s...@0/s...@3,0: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@8: file changed as we read
it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@8/p...@0: file changed as
we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@8/p...@0/p...@1: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@8/p...@0/p...@2: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@8/p...@0/p...@8: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@8/p...@0/p...@9: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@8/p...@0/p...@a: file
changed as we read it
? /usr/local/bin/tar: ./devices/p...@0/p...@0/p...@9: file changed
SV: amvault report - *** THE DUMPS DID NOT FINISH PROPERLY!
ok then I must overide the tapetype def with -o option in the amvault command. No, it needs to be in the tapetype. It's weird, I know, but there are some difficult problems with combining those two sections. Thanks Gunnar Gunnarsson
SV: amvault report - *** THE DUMPS DID NOT FINISH PROPERLY!
Snapshot is created but not used see below in 3.2.0beta3 Sat Oct 9 22:23:08 2010: amgtar: Spawning /usr/local/bin/tar /usr/local/bin/tar --create --verbose --file - --directory / --one-file-system --no-check-device --listed-incremental /var/opt/lib/amanda/gnutar-lists/hansabck_1.new --sparse --ignore-failed-read --totals . in pipeline but in 3.1.1 snapshot is created and used - the config is the same. Sat Oct 9 20:10:26 2010: amgtar: Spawning /usr/local/bin/tar /usr/local/bin/tar --create --verbose --file - --directory //.zfs/snapshot/amanda-_-current --one-file-system --no-check-device --listed-incremental/var/opt/lib/amanda/gnutar-lists/tx5019__0.new --sparse --ignore-failed-read --totals . in pipeline ok it's different machines but I downgraded the client and the log was overwritten. Hmm, I don't think we made any changes to how GNU Tar is invoked that would cause it to no longer be able to read special device files - it never could do that. If you still have a sendbackup debug log from 3.1.x, can you compare the tar invocation to the one made by 3.2.0? Thanks Gunnar Gunnarsson
SV: Testing amvault
Hi Dustin, Good work releasing beta3. Few thing that come to my mind while testing amvault regarding slections of dumps. - latest dump with --src-timestamp latest - all dumps without specifying --src-timestamp - parts by specifying dates eg. --src-timestamp 20100926-20101008 I would likte to select all but only the most recent version. Thanks Gunnar Gunnarsson On Tue, Oct 5, 2010 at 1:11 AM, Gunnarsson, Gunnar gunnar.gunnars...@svk.se wrote: I can do some vault testing later this week. What version should I test ? Is the issue with large logfiles resolved ? With luck, we'll have a 3.2.0beta2 out by Thursday that will resolve this problem. If not, you can apply http://github.com/zmanda/amanda/commit/cfe0f6e656401da07d3c9040a763055808185285.patch in place (don't worry about the manpage hunk - just patch /usr/sbin/amvault and $amperldir/Amanda/Taper/Scribe.pm). Dustin -- Open Source Storage Engineer http://www.zmanda.com
amvault - version 3.2.0beta3
I'm vaulting from config hansa and I'm getting parse error about hansa-vault config ? parse error: could not open conf file /etc/opt/amanda/hansa-vault/amanda.conf: No such file or directory amdevcheck: errors processing config file at /opt/sfw/sbin/amdevcheck line 113. parse error: could not open conf file /etc/opt/amanda/hansa-vault/amanda.conf: No such file or directory amdevcheck: errors processing config file at /opt/sfw/sbin/amdevcheck line 113. I notided that autolabel any is not allowed - went for volume-error instead. Thanks Gunnar Gunnarsson
SV: Testing amvault
ok this will vault all full dumps in my configuation but can I select only the most resent for each filesystem ? amvault --dry-run --fulls-only --dst-changer vault \ --label-template EMS1-7% --autolabel volume-error hansa What does most recent version mean in this case? Thanks Gunnar Gunnarsson
amvault report - *** THE DUMPS DID NOT FINISH PROPERLY!
Is this because amvault is considered a flush ?
Tape usage statistic is rather odd as well.
Otherwise is seems to work quite well - good work !
Thanks Gunnar Gunnarsson
*** THE DUMPS DID NOT FINISH PROPERLY!
The dumps were vaulted to tape EMS1-71.
The next 2 tapes Amanda expects to use are: EMS1-13, EMS1-14.
STATISTICS:
Total Full Incr. Level:#
Estimate Time (hrs:min) 0:00
Run Time (hrs:min) 1:59
Dump Time (hrs:min) 0:00 0:00 0:00
Output Size (meg)0.00.00.0
Original Size (meg) 0.00.00.0
Avg Compressed Size (%) -- -- --
DLEs Dumped0 0 0
Avg Dump Rate (k/s) -- -- --
Tape Time (hrs:min) 1:59 1:59 0:00
Tape Size (meg) 261950.7 261950.70.0
Tape Used (%) 532.9 532.90.0
DLEs Taped 462462 0
Parts Taped 462462 0
Avg Tp Write Rate (k/s) 37421.537421.5--
USAGE BY TAPE:
Label Time Size % DLEs Parts
EMS1-71 1:59 261951M 532.9 462 462
SV: amvault report - *** THE DUMPS DID NOT FINISH PROPERLY!
No I was wrong about it only delayed the flushing until at the end. Is that a new behaviour in versioh 3.2.0 ? The only problem I got was on the client side for the root filesystem zfs. /-- hansabck / lev 1 FAILED [/usr/local/bin/tar exited with status 2: see /tmp/amanda/client/hansa/amgtar.20101008223221000.debug] sendbackup: info BACKUP=APPLICATION sendbackup: info APPLICATION=amgtar sendbackup: info RECOVER_CMD=/usr/bin/gzip -dc |/opt/sfw/libexec/amanda/application/amgtar restore [./file-to-restore]+ sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end ? /usr/local/bin/tar: ./devices/ramdisk-root: Cannot savedir: No such file or directory ? /usr/local/bin/tar: ./devices/ramdisk-root/: Warning: Cannot stat: No such file or directory -- GG Let's take this back on the list.
amrecover verson 3.1.2
Hi, I'm using the the poor mans changer-manual on one system and amrecover is not so happy with it. The setup works with andump. Thanks Gunnar Gunnarsson amanda.conf tpchanger chg-manual # the tape-changer glue script changerfile /etc/opt/amanda/hansa-arkive/changer tapedev tape:/dev/rmt/0n From log file Mon Oct 4 11:38:22 2010: amidxtaped: find_volume labeled 'HANSA-ARKIVE-11' Mon Oct 4 11:38:22 2010: amidxtaped: parse_inventory: load slot 0 Mon Oct 4 11:38:22 2010: amidxtaped: warning: chg-manual uses deprecated device naming convention; using tape:chg-manual instead. Mon Oct 4 11:38:22 2010: amidxtaped: Device chg-manual error = 'Can't open tape device chg-manual: No such file or directory' Mon Oct 4 11:38:22 2010: amidxtaped: Device chg-manual setting status flag(s): DEVICE_STATUS_DEVICE_ERROR Mon Oct 4 11:38:22 2010: amidxtaped: new Amanda::Changer::Error: type='fatal', message='Can't open tape device chg-manual: No such file or directory' Mon Oct 4 11:38:22 2010: amidxtaped: CTL MESSAGE Can't open tape device chg-manual: No such file or directory Mon Oct 4 11:38:22 2010: amidxtaped: CTL FEEDME HANSA-ARKIVE-11 amrecover extract Extracting files using tape drive chg-manual on host localhost. The following tapes are needed: HANSA-ARKIVE-11 Extracting files using tape drive chg-manual on host localhost. Load tape HANSA-ARKIVE-11 now Continue [?/Y/n/s/d]? Y Can't open tape device chg-manual: No such file or directory Load tape HANSA-ARKIVE-11 now Continue [?/Y/n/d]?
SV: amreport version 3.1.2
Yes I hope so. What version should I go for ? Thanks Gunnar Gunnarsson Incidentally, I can't help but notice that you're backing up disks named /var/amanda/amandatapes/ems/slotNN. Will amvault be able to help you out? If so, could you find some time to test it before I set sail in a week and a half?
amreport version 3.1.2
Hi, I upgraded Amanda version from 2.6.1 to 3.1.2 and amreport reports: FAILURE DUMP SUMMARY: hansabck /var/amanda/amandatapes/ems1/slot13 lev 0 partial taper: No space left on device hansabck /var/amanda/amandatapes/ems1/slot65 lev 0 partial taper: No space left on device But they are written on the next tape, I'm not using tape spanning. hansabck -ms1/slot13 0 18409 18409-- PARTIAL 865:55 362.8 hansabck -ms1/slot65 0 25100 25100-- PARTIAL 31:30 13599.1 Thanks Gunnar Gunnarsson
SV: amreport version 3.1.2
It is reported as failed see below and those parts are added twice - filesystem 66 parts 68. In earlier version it is reported correctly. These dumps were to tapes HANSA-ARKIVE-9, HANSA-ARKIVE-10, HANSA-ARKIVE-11. The next 4 tapes Amanda expects to use are: HANSA-ARKIVE-12, HANSA-ARKIVE-13, HANSA-ARKIVE-14, HANSA-ARKIVE-15. FAILURE DUMP SUMMARY: hansabck /var/amanda/amandatapes/ems1/slot13 lev 0 partial taper: No space left on device hansabck /var/amanda/amandatapes/ems1/slot65 lev 0 partial taper: No space left on device Estimate Time (hrs:min) 0:01 Run Time (hrs:min) 25:51 Dump Time (hrs:min)34:33 34:33 0:00 Output Size (meg) 1292329.1 1292329.10.0 Original Size (meg)1292329.1 1292329.10.0 Avg Compressed Size (%)100.0 100.0-- Filesystems Dumped66 66 0 Avg Dump Rate (k/s) 10639.110639.1-- Tape Time (hrs:min)22:38 22:38 0:00 Tape Size (meg)1248820.2 1248820.20.0 Tape Used (%) 359.6 359.60.0 Filesystems Taped 68 68 0 Parts Taped 68 68 0 Avg Tp Write Rate (k/s) 15690.715690.7-- USAGE BY TAPE: Label Time Size %NbNc HANSA-ARKIVE-9 2:45 430321M 111.52525 HANSA-ARKIVE-10 2:50 440969M 114.22424 HANSA-ARKIVE-11 2:22 398970M 103.31919 OK .. so what's the problem? Thanks Gunnar Gunnarsson
SV: Nitpick: scripts and pfexec/sudo
Yes this with Script API will do the work, I overlooked the Script API. Perhaps it should be better documented ? #! /bin/sh pfexec --pfexec-params /path/to/my/script
SV: Nitpicks?
Hi Dustin,
A script-tool application for running script taking parameters such as path to
script, PFEXEC or SUDO and when to run etc
define script-tool amplugin-script {
plugin amplugin-script
property script path to script to run
property PFEXEC YES
property SUDO no
execute-on pre-dle-backup,pre-dle-amcheck,post-dle-backup
}
Tailored backup script could then more easily being intergraded into amanda
dump types.
Thanks Gunnar Gunnarsson
I bet most of you have some small nitpick with Amanda that you've never felt
warranted an email. Well, now's your chance! I'd like to put some polish on
Amanda, and it's hard for me to see the areas that need burnishing, since I
work on Amanda all day, every day.
- typo in a manpage?
- command-line usage oddity?
- confusing use of terminology?
- something else?
SV: lvm snapshots
Hi,
I'm backing up active filesystem using ufs snapsshot and metaonline/offline for
mirrors in Solaris.
Hopefully this gives you some ideas how to proceed.
define script-tool fssnap-script {
comment snapshot wrapper
plugin fssnap-script
property FSSNAP-WRAPPER /etc/opt/amanda/metaonline_snap.sh
execute_on pre-dle-backup,pre-dle-amcheck,post-dle-backup
execute_where client # or client
}
define dumptype snap-dump {
comp-root
compress client custom
client_custom_compress /usr/local/bin/pigz
script fssnap-script
auth bsdtcp
}
I am thinking of backing up some DLEs that are fairly active from a snapshot
rather than from the filesystem directly. I looked for documentation on doing
this but could find none, nor any HowTo's.
SV: ufsdump on submirror
Yes it worked out well in the end. I had one Solaris issue that ufsdump complained about not being able to create some directory see below but this disappeared when file system was mounted. I guess it's some kind of security issue as backup is run not by root. sendbackup: 115: strange(?): Unable to create temporary directory in any of the directories listed below: sendbackup: 115: strange(?): /tmp/ sendbackup: 115: strange(?): /var/tmp/ sendbackup: 115: strange(?): / sendbackup: 115: strange(?): Please correct this problem and rerun the program. I believe the short answer is yes. We had alternate boot roots on a FW system and used ufsdump to dump the unmounted boot roots (as well as the active boot root). Since it was umounted as specified the device string in the DLE, /dev/dsk/cX... On Wed, Jun 30, 2010 at 03:37:03PM +0200, Gunnarsson, Gunnar wrote: Hi, I'm backuping a busy UFS filesystem in Solaris, fssnap can't be used because it's not supported in SUN HA Cluster. so I'm stuck with using submirrors and taken them offline/online. I do this using amgtar and script tool. My question is can this be done using dump instead. I'm hoping that ufsdump is performing better than gtar on this filesystem - more than 2 millions files. Thanks Gunnar Gunnarsson --- Brian R Cuttler brian.cutt...@wadsworth.org Computer Systems Support(v) 518 486-1697 Wadsworth Center(f) 518 473-6384 NYS Department of HealthHelp Desk 518 473-0773 IMPORTANT NOTICE: This e-mail and any attachments may contain confidential or sensitive information which is, or may be, legally privileged or otherwise protected by law from further disclosure. It is intended only for the addressee. If you received this in error or from someone who was not authorized to send it to you, please do not distribute, copy or use it or any attachments. Please notify the sender immediately by reply e-mail and delete this from your system. Thank you for your cooperation.
Strange dump summary
Dump gives a strange dump summary if file system is not mounted - is this required or a bug ? To illustrate in my disk list: tx5019 /mnt /dev/lofi/2 comp-root STRANGE DUMP DETAILS: /-- tx5019 /mnt lev 1 STRANGE sendbackup: start [tx5019:/mnt level 1] sendbackup: info BACKUP=/usr/sbin/ufsdump sendbackup: info RECOVER_CMD=/usr/bin/gzip -dc |/usr/sbin/ufsrestore -xpGf - ... sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end ? Unable to create temporary directory in any of the directories listed below: ? /tmp/ ? /var/tmp/ ? / ? Please correct this problem and rerun the program. | DUMP: Date of this level 1 dump: Mon Jul 19 13:40:30 2010 | DUMP: Date of last level 0 dump: Mon Jul 19 13:36:28 2010 | DUMP: Dumping /dev/rlofi/2 to standard output. | DUMP: Mapping (Pass I) [regular files] | DUMP: Mapping (Pass II) [directories] | DUMP: Writing 32 Kilobyte records | DUMP: Estimated 326 blocks (163KB) on 0.00 tapes. | DUMP: Dumping (Pass III) [directories] | DUMP: Dumping (Pass IV) [regular files] | DUMP: 190 blocks (95KB) on 1 volume at 9500 KB/sec | DUMP: DUMP IS DONE | DUMP: Level 1 dump on Mon Jul 19 13:40:30 2010 sendbackup: size 95 sendbackup: end If file system is mounted dump works as expexted. Thanks Gunnar Gunnarsson
SV: upgrade to 3.1.1
The issue is resolved it needs to have write permission in the sysconfdir directory as well. I my case its /etc/opt/amanda which this differs from previous versions. Thanks Gunnar Gunnarsson -Ursprungligt meddelande- Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell Skickat: den 2 juli 2010 23:56 Till: Gunnarsson, Gunnar Kopia: amanda-users@amanda.org Ämne: Re: upgrade to 3.1.1 On Fri, Jul 2, 2010 at 4:34 PM, Gunnarsson, Gunnar gunnar.gunnars...@svk.se wrote: Perms are okay, it was working for versions 3.1.0. # ls -l /etc/opt/amanda/orabck/tapelist -rw-r--r-- 1 amanda amanda 259 Jun 28 20:26 /etc/opt/amanda/orabck/tapelist How about for the enclosing directory? Amanda tries to write the tapelist safely by writing to a temporary file in the same directory and renaming. Is the taper running as 'amanda' or 'amandabackup'? Dustin -- Open Source Storage Engineer http://www.zmanda.com
sendbackup strange ? amanda version 3.1.1
See below strange listing from senbackup unable to create temporary direcotry
... etc.
What's going on what does this mean ? See debug listing below.
Thanks Gunnar Gunnarsson
Sun Jul 4 20:06:57 2010: sendbackup: pid 27319 ruid 91 euid 91 version 3.1.1:
start at Sun Jul 4 20:06:57 2010 Sun Jul 4 20:06:57 2010: sendbackup: Version
3.1.1 Sun Jul 4 20:06:57 2010: sendbackup: pid 27319 ruid 91 euid 91 version
3.1.1: rename at Sun Jul 4 20:06:57 2010
Sun Jul 4 20:06:57 2010: sendbackup: Parsed request as: program `DUMP'
Sun Jul 4 20:06:57 2010: sendbackup: disk `shared'
Sun Jul 4 20:06:57 2010: sendbackup: device
`/dev/md/ediprod/rdsk/d83'
Sun Jul 4 20:06:57 2010: sendbackup: level 1
Sun Jul 4 20:06:57 2010: sendbackup: since NODATE
Sun Jul 4 20:06:57 2010: sendbackup: options `'
Sun Jul 4 20:06:57 2010: sendbackup: datapath `AMANDA'
Sun Jul 4 20:06:57 2010: sendbackup: Spawning
/opt/sfw/libexec/amanda/application/fssnap-script fssnap-script PRE-DLE-BACKUP
--execute-where client --config daily --host x5045 --disk shared --device
/dev/md/ediprod/rdsk/d83 --level 1 --fssnap-wrapper
/etc/opt/amanda/synchrony_snap.sh in pipeline Sun Jul 4 20:07:24 2010:
sendbackup: start: x5045:shared lev 1 Sun Jul 4 20:07:24 2010: sendbackup:
Spawning /usr/bin/gzip /usr/bin/gzip --fast in pipeline Sun Jul 4 20:07:24
2010: sendbackup: dump: pid 27426: /usr/bin/gzipSun Jul 4 20:07:24 2010:
sendbackup: --fastSun Jul 4 20:07:24 2010: sendbackup:
Sun Jul 4 20:07:24 2010: sendbackup: dumping device '/dev/md/ediprod/rdsk/d83'
with ''
Sun Jul 4 20:07:24 2010: sendbackup: Spawning /usr/sbin/ufsdump dump 1usf
1048576 - /dev/md/ediprod/rdsk/d83 in pipeline Sun Jul 4 20:07:24 2010:
sendbackup: Started backup Sun Jul 4 20:07:24 2010: sendbackup: Started index
creator: /usr/sbin/ufsrestore -tvf - 21 | sed -e '
s/^leaf[]*[0-9]*[ ]*\.//
t
/^dir[ ]/ {
s/^dir[ ]*[0-9]*[ ]*\.//
s%$%/%
t
}
d
'
Sun Jul 4 20:07:24 2010: sendbackup: 115: strange(?): Unable to create
temporary directory in any of the directories listed below:
Sun Jul 4 20:07:24 2010: sendbackup: 115: strange(?): /tmp/
Sun Jul 4 20:07:24 2010: sendbackup: 115: strange(?): /var/tmp/
Sun Jul 4 20:07:24 2010: sendbackup: 115: strange(?): /
Sun Jul 4 20:07:24 2010: sendbackup: 115: strange(?): Please correct this
problem and rerun the program.
Sun Jul 4 20:07:25 2010: sendbackup: 90: normal(|): DUMP: Date of this
level 1 dump: Sun Jul 04 20:07:25 2010
Sun Jul 4 20:07:25 2010: sendbackup: 90: normal(|): DUMP: Date of last
level 0 dump: Thu Jul 01 20:17:24 2010
Sun Jul 4 20:07:25 2010: sendbackup: 90: normal(|): DUMP: Dumping
/dev/md/ediprod/rdsk/d83 to standard output.
Sun Jul 4 20:08:54 2010: sendbackup: 90: normal(|): DUMP: Mapping (Pass I)
[regular files]
Sun Jul 4 20:09:50 2010: sendbackup: 90: normal(|): DUMP: Mapping (Pass
II) [directories]
Sun Jul 4 20:13:24 2010: sendbackup: 90: normal(|): DUMP: Mapping (Pass
II) [directories]
Sun Jul 4 20:13:25 2010: sendbackup: 90: normal(|): DUMP: Mapping (Pass
II) [directories]
Sun Jul 4 20:13:26 2010: sendbackup: 90: normal(|): DUMP: Mapping (Pass
II) [directories]
Sun Jul 4 20:13:27 2010: sendbackup: 90: normal(|): DUMP: Mapping (Pass
II) [directories]
Sun Jul 4 20:13:35 2010: sendbackup: 90: normal(|): DUMP: Writing 32
Kilobyte records
Sun Jul 4 20:13:35 2010: sendbackup: 90: normal(|): DUMP: Estimated
65086486 blocks (31780.51MB) on 0.47 tapes.
Sun Jul 4 20:13:36 2010: sendbackup: 90: normal(|): DUMP: Dumping (Pass
III) [directories]
Sun Jul 4 20:13:57 2010: sendbackup: 90: normal(|): DUMP: Dumping (Pass
IV) [regular files]
Sun Jul 4 20:23:35 2010: sendbackup: 90: normal(|): DUMP: 10.92% done,
finished in 1:21
Sun Jul 4 20:33:35 2010: sendbackup: 90: normal(|): DUMP: 23.43% done,
finished in 1:05
Sun Jul 4 20:43:35 2010: sendbackup: 90: normal(|): DUMP: 38.04% done,
finished in 0:48
Sun Jul 4 20:53:35 2010: sendbackup: 90: normal(|): DUMP: 51.75% done,
finished in 0:37
Sun Jul 4 21:03:35 2010: sendbackup: 90: normal(|): DUMP: 66.07% done,
finished in 0:25
Sun Jul 4 21:13:35 2010: sendbackup: 90: normal(|): DUMP: 90.88% done,
finished in 0:06
Sun Jul 4 21:19:13 2010: sendbackup: 54:size(|): DUMP: 65071166 blocks
(31773.03MB) on 1 volume at 8262 KB/sec
Sun Jul 4 21:19:13 2010: sendbackup: 90: normal(|): DUMP: DUMP IS DONE
Sun Jul 4 21:19:13 2010: sendbackup: 90: normal(|): DUMP: Level 1 dump on
Sun Jul 04 20:07:25 2010
Sun Jul 4 21:19:13 2010: sendbackup: Index created successfully Sun Jul 4
21:19:18 2010: sendbackup: Parsed backup messages Sun Jul 4 21:19:18 2010:
sendbackup: Spawning /opt/sfw/libexec/amanda/application/fssnap-script
fssnap-script POST-DLE-BACKUP --execute-where client --config daily
Strange sendbackup message ufsdump
Hi,
See below message from sendbackup, I'm running version 3.1.1 Solaris 10 sparc.
What should correct - /tmp and /var/tmp are writable for others ?
Thanks Gunnar Gunnarsson
Fri Jul 2 20:07:33 2010: sendbackup: dumping device '/dev/md/ediprod/rdsk/d83'
with ''
Fri Jul 2 20:07:33 2010: sendbackup: Spawning /usr/sbin/ufsdump dump 1usf
1048576 - /dev/md/
ediprod/rdsk/d83 in pipeline
Fri Jul 2 20:07:33 2010: sendbackup: Started backup
Fri Jul 2 20:07:33 2010: sendbackup: Started index creator:
/usr/sbin/ufsrestore -tvf - 21
| sed -e '
s/^leaf[]*[0-9]*[ ]*\.//
t
/^dir[ ]/ {
s/^dir[ ]*[0-9]*[ ]*\.//
s%$%/%
t
}
d
'
Fri Jul 2 20:07:33 2010: sendbackup: 115: strange(?): Unable to create
temporary directory in any of the directories listed below:
Fri Jul 2 20:07:33 2010: sendbackup: 115: strange(?): /tmp/
Fri Jul 2 20:07:33 2010: sendbackup: 115: strange(?): /var/tmp/
Fri Jul 2 20:07:33 2010: sendbackup: 115: strange(?): /
Fri Jul 2 20:07:33 2010: sendbackup: 115: strange(?): Please correct this
problem and rerun the program.
upgrade to 3.1.1
Hi, I got this error while I upgraded from 3.1.0beta2 to 3.1.1 ? File is ownered by the backup user. The next tape Amanda expects to use is: 1 new tape. FAILURE DUMP SUMMARY: taper: FATAL Could not open '/etc/opt/amanda/orabck/tapelist' for writing: Permission denied at /usr/perl5/site_perl/5.8.4/Amanda/Tapelist.pm line 280. Thanks Gunnar Gunnarsson
SV: upgrade to 3.1.1
Perms are okay, it was working for versions 3.1.0. # ls -l /etc/opt/amanda/orabck/tapelist -rw-r--r-- 1 amanda amanda 259 Jun 28 20:26 /etc/opt/amanda/orabck/tapelist -Ursprungligt meddelande- Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell Skickat: den 2 juli 2010 22:59 Till: Gunnarsson, Gunnar Kopia: amanda-users@amanda.org Ämne: Re: upgrade to 3.1.1 On Fri, Jul 2, 2010 at 2:44 PM, Gunnarsson, Gunnar gunnar.gunnars...@svk.se wrote: The next tape Amanda expects to use is: 1 new tape. FAILURE DUMP SUMMARY: taper: FATAL Could not open '/etc/opt/amanda/orabck/tapelist' for writing: Permission denied at /usr/perl5/site_perl/5.8.4/Amanda/Tapelist.pm line 280. Hmm - did you check the perms on that file? Dustin -- Open Source Storage Engineer http://www.zmanda.com
SV: upgrade to 3.1.1
It's own by the same user and taper is running as the user amanda. Fri Jul 2 19:30:05 2010: taper: pid 29208 ruid 91 euid 91 version 3.1.1: start at Fri Jul 2 19:30:05 2010 Fri Jul 2 19:30:05 2010: taper: pid 29208 ruid 91 euid 91 version 3.1.1: rename at Fri Jul 2 19:30:05 2010 # ls -ld /etc/opt/amanda/orabck drwxr-xr-x 2 amanda amanda 8 May 28 07:53 /etc/opt/amanda/orabck # ls -ldn /etc/opt/amanda/orabck drwxr-xr-x 2 91 91 8 May 28 07:53 /etc/opt/amanda/orabck -Ursprungligt meddelande- Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell Skickat: den 2 juli 2010 23:56 Till: Gunnarsson, Gunnar Kopia: amanda-users@amanda.org Ämne: Re: upgrade to 3.1.1 On Fri, Jul 2, 2010 at 4:34 PM, Gunnarsson, Gunnar gunnar.gunnars...@svk.se wrote: Perms are okay, it was working for versions 3.1.0. # ls -l /etc/opt/amanda/orabck/tapelist -rw-r--r-- 1 amanda amanda 259 Jun 28 20:26 /etc/opt/amanda/orabck/tapelist How about for the enclosing directory? Amanda tries to write the tapelist safely by writing to a temporary file in the same directory and renaming. Is the taper running as 'amanda' or 'amandabackup'? Dustin -- Open Source Storage Engineer http://www.zmanda.com
ufsdump on submirror
Hi, I'm backuping a busy UFS filesystem in Solaris, fssnap can't be used because it's not supported in SUN HA Cluster. so I'm stuck with using submirrors and taken them offline/online. I do this using amgtar and script tool. My question is can this be done using dump instead. I'm hoping that ufsdump is performing better than gtar on this filesystem - more than 2 millions files. Thanks Gunnar Gunnarsson
SV: default dumptype DUMP
Gunnarsson, Gunnar wrote: Does DUMP dumptype have any properties that can be redefined ? I need to follwing task: stop the application, dettach the mirror, ufsdump attach the mirror This can be done with gtar but can this be done with native backup tools as well ? I Hoogendyk, Chris wrote: Haven't seen an answer to this yet, so I'll hazard a comment and question. One approach is to do a wrapper, as in http://wiki.zmanda.com/index.php/Backup_client#Chris_Hoogendyk.27s_Example, where I'm doing an fssnap before the ufsdump and following it with deleting the snapshot. You could modify it to fit your needs. That wrapper also has a stop application for xntpd (which is a real time application) when doing an fssnap of the root partition because of a bug in fssnap. What version are you using? I'm using 2.6.* but I'm upgrading to 3.1.0. I'm running Solaris 10 and I can't use fssnap for UFS becuase it's not supported in a cluster evironment. ZFS didn't work out either because it is using variable block size in the filesystem which breaks part of the application so the last resort is using a 3-way mirror and doing detach stopping the application and then restart it, go through with backup and then detach the third-mirror and resync. What are the improvement for scripting in version 3.1.0 and how can this be wrapped into the client ? Since the filesystem contians a lot of files ( few millions ) I belive that ufsdump works better than gtar. Thanks Gunnar Gunnarsson
default dumptype DUMP
Hi, Does DUMP dumptype have any properties that can be redefined ? I need to follwing task: stop the application, dettach the mirror, ufsdump attach the mirror This can be done with gtar but can this be done with native backup tools as well ? I Thanks Gunnar Gunnarsson
amrecover core dumps Amanda 3.1.0beta2
While testing amrecover I got core dump see below on Solaris sparc amrecover extract Extracting files from holding disk on host localhost. The following files are needed: /var/opt/amanda/hd1/20100527211005/generisgaspr0 Extracting from file /var/opt/amanda/hd1/20100527211005/generisgasproddb._orac0 Restoring files into directory /var/tmp/backup All existing files in /var/tmp/backup can be deleted Continue [?/Y/n]? ? Enter yes to continue, no to stop Continue [?/Y/n]? n Bus Error (core dumped) Extracting files does work as expected for this dumptype at least. Thanks Gunnar
New options autolabel Amanda 3.1.0bet2
I got the fatal error see below while testing the new option autolabel, the configuration was empty and amcheck -s was ok. The next tape Amanda expects to use is: 1 new tape. FAILURE DUMP SUMMARY: taper: FATAL Can't call method lookup_tapelabel on an undefined value at /opt/sfw/libexec/amanda/taper line 334. Thu May 27 21:10:09 2010: taper: Device tape:/dev/rmt/0n error = 'Error readingAmanda header: Error reading 32768 bytes from /dev/rmt/0n: I/O error' Thu May 27 21:10:09 2010: taper: Device tape:/dev/rmt/0n setting status flag(s): DEVICE_STATUS_DEVICE_ERROR, DEVICE_STATUS_VOLUME_UNLABELED, and DEVICE_STATUS_VOLUME_ERROR Thu May 27 21:10:09 2010: taper: sl48: setting current slot to 11 Thu May 27 21:10:09 2010: taper: Amanda::Taper::Scan::traditional result: 'Orabck-Unix-001' on tape:/dev/rmt/0n slot 11, mode 2 Thu May 27 21:34:39 2010: taper: Amanda::Taper::Scribe setting up a transfer with split method none Thu May 27 21:34:39 2010: taper: Starting x...@8d3d88 (xfersourcehold...@8cf420 - xferdesttapersplit...@8d7028) Thu May 27 21:34:39 2010: taper: Final linkage: xfersourcehold...@8cf420 -(PULL_BUFFER)- xferelementg...@8d8830 -(PUSH_BUFFER)- xferdesttapersplit...@8d7028 Thu May 27 21:34:39 2010: taper: Building type TAPESTART header of 32768-32768 bytes with name='Orabck-Unix-001' disk='' dumplevel=0 and blocksize=32768 Thu May 27 21:34:45 2010: taper: critical (fatal): Can't call method lookup_tapelabel on an undefined value at /opt/sfw/libexec/amanda/taper line 334. Thanks Gunnar
Using flush threshold
Should I get While using threshold should I get a tape errors ? flush-threshold-dumped 90 flush-threshold-scheduled 90 taperflush 90 autoflush yes *** A TAPE ERROR OCCURRED: [flush-threshold-dumped criteria not met]. There are 65959M of dumps left in the holding disk. They will be flushed on the next run. The next tape Amanda expects to use is: 1 new tape. The next new tape already labelled is: Orabck-Unix-001. FAILURE DUMP SUMMARY: generisgasproddb /oracle/backup/EDMSGAS lev 0 FLUSH flush-threshold-dumped criteria not met
Using flush threshold
How come I'm getting tape error while using flush threshod ? *** A TAPE ERROR OCCURRED: [flush-threshold-dumped criteria not met]. There are 65959M of dumps left in the holding disk. They will be flushed on the next run. flush-threshold-dumped 90 flush-threshold-scheduled 90 taperflush 90 autoflush yes Thanks Gunnar
SV: amtapetype Amanda 3.1.0beta2
Hi, I've tested the tape drivers and they seems to be working fine, how does drivers detect end of tape. This seems to be a specific Solaris issue ? What do you mean by re-creating the device ? Thanks Gunnar -Ursprungligt meddelande- Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell Skickat: den 24 maj 2010 16:46 Till: Gunnarsson, Gunnar Kopia: amanda-users@amanda.org Ämne: Re: amtapetype Amanda 3.1.0beta2 On Sat, May 22, 2010 at 5:46 AM, Gunnarsson, Gunnar gunnar.gunnars...@svk.se wrote: Wrote 406883762176 bytes at 7552 kb/sec Writing smaller files (4068835328 bytes) to determine filemark. amtapetype: Error writing label 'amtapetype-2046333855': Error writing block: Mysterious short write on tape device: Tried 32768, got 0 at /opt/sfw/sbin/amtapetype line 89. Ah, it's a similar problem as in the previous case, then: the device goes into an error state, and that state sticks. Probably the better solution is to re-create the device as necessary. Dustin -- Open Source Storage Engineer http://www.zmanda.com
SV: amtapetype Amanda 3.1.0beta2
Hi Dustin,
I tested amtapetype with version 2.6.1p1 with same HW. It goes through the
compressions test in the beginning where I got
error in the Amanda 3.1.0beta2 version, but then when it comes to end of tape
the driver returns DEVICE_STATUS_DEVICE_ERROR
with causes short write on tape device. But see below what mt -f status returns
- EOT.
Thanks Gunnar
bash-3.00# /opt/sfw/sbin/amtapetype -f tape:/dev/rmt/0n
Applying heuristic check for compression.
Wrote random (uncompressible) data at 9076215.87301587 bytes/sec
Wrote fixed (compressible) data at 38120106.667 bytes/sec
Compression: enabled
Writing one file to fill the volume.
Wrote less than 100MB to the device: Error writing block: Error 0
# mt -f /dev/rmt/0n status
HP Ultrium LTO 3 tape drive:
sense key(0x13)= EOT residual= 0 retries= 0
file no= 2 block no= 0
1274535946.994402: amtapetype: Final linkage: xfersourceran...@438858
-(PULL_BUFFER)-xferdestdev...@4394b0
1274578788.428012: amtapetype: Device /dev/rmt/0n error = 'Mysterious short
write on tape device: Tried 32768, got 0'
1274578788.458837: amtapetype: Device /dev/rmt/0n setting status flag(s):
DEVICE_STATUS_DEVICE_ERROR
1274578788.459176: amtapetype: Device /dev/rmt/0n error = 'Error writing
block:Error 0'
1274578788.459231: amtapetype: Device /dev/rmt/0n setting status flag(s):
DEVICE_STATUS_DEVICE_ERROR
1274578788.498811: amtapetype: Cancelling x...@3533f0
(xfersourceran...@438858 - xferdestdev...@4394b0)
-Ursprungligt meddelande-
Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell
Skickat: den 21 maj 2010 16:44
Till: Gunnarsson, Gunnar
Kopia: amanda-users@amanda.org
Ämne: Re: amtapetype Amanda 3.1.0beta2
Ah - after the device has encountered an error, it won't give properties
anymore (case doesn't matter). Try moving the block_size up in the script:
diff --git a/device-src/amtapetype.pl b/device-src/amtapetype.pl index
5bd7411..80c0ad7 100644
--- a/device-src/amtapetype.pl
+++ b/device-src/amtapetype.pl
@@ -117,6 +117,9 @@ sub write_one_file(%) {
my $pattern = $options{'PATTERN'} || 'FIXED';
my $max_time = $options{'MAX_TIME'} || 0;
+# get the block size now, while the device is still working
+my $block_size = $device-property_get(block_size);
+
# start the device
my $hdr = Amanda::Header-new();
$hdr-{type} = $Amanda::Header::F_DUMPFILE; @@ -182,7 +185,6 @@ sub
write_one_file(%) {
# OK, we finished, update statistics (even if we saw an error)
my $blocks_written = $device-block();
-my $block_size = $device-property_get(block_size);
$stats-{$pattern}-{BYTES} += $blocks_written * $block_size;
$stats-{$pattern}-{FILES} += 1;
$stats-{$pattern}-{TIME} += $duration;
Let me know if that helps? Thanks for your patience!
Dustin
--
Open Source Storage Engineer
http://www.zmanda.com
SV: amtapetype Amanda 3.1.0beta2
The driver reports DEVICE_STATUS_DEVICE_ERROR while see below the mt command
reports EOT.
-- GG
$ amtapetype -f /dev/rmt/0n
Applying heuristic check for compression.
blocksize = 32768 blocks written = 14176
blocksize = 32768 blocks written = 14176
Wrote random (uncompressible) data at 7492244.64516129 bytes/sec
Wrote fixed (compressible) data at 35732243.6923077 bytes/sec
Compression: enabled
Writing one file to fill the volume.
blocksize = 32768 blocks written = 12417107
Wrote 406883762176 bytes at 7552 kb/sec
Writing smaller files (4068835328 bytes) to determine filemark.
amtapetype: Error writing label 'amtapetype-2046333855': Error writing block:
Mysterious short write on tape device: Tried 32768, got 0 at
/opt/sfw/sbin/amtapetype line 89.
$ mt -f /dev/rmt/0n status
HP Ultrium LTO 3 tape drive:
sense key(0x13)= EOT residual= 0 retries= 0
file no= 2 block no= 0
-Ursprungligt meddelande-
Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell
Skickat: den 21 maj 2010 16:44
Till: Gunnarsson, Gunnar
Kopia: amanda-users@amanda.org
Ämne: Re: amtapetype Amanda 3.1.0beta2
Ah - after the device has encountered an error, it won't give properties
anymore (case doesn't matter). Try moving the block_size up in the script:
diff --git a/device-src/amtapetype.pl b/device-src/amtapetype.pl index
5bd7411..80c0ad7 100644
--- a/device-src/amtapetype.pl
+++ b/device-src/amtapetype.pl
@@ -117,6 +117,9 @@ sub write_one_file(%) {
my $pattern = $options{'PATTERN'} || 'FIXED';
my $max_time = $options{'MAX_TIME'} || 0;
+# get the block size now, while the device is still working
+my $block_size = $device-property_get(block_size);
+
# start the device
my $hdr = Amanda::Header-new();
$hdr-{type} = $Amanda::Header::F_DUMPFILE; @@ -182,7 +185,6 @@ sub
write_one_file(%) {
# OK, we finished, update statistics (even if we saw an error)
my $blocks_written = $device-block();
-my $block_size = $device-property_get(block_size);
$stats-{$pattern}-{BYTES} += $blocks_written * $block_size;
$stats-{$pattern}-{FILES} += 1;
$stats-{$pattern}-{TIME} += $duration;
Let me know if that helps? Thanks for your patience!
Dustin
--
Open Source Storage Engineer
http://www.zmanda.com
SV: amtapetype Amanda 3.1.0beta2
See below the blocksize in the third print out is defined !
# grep -i block_size /opt/sfw/sbin/amtapetype | grep proper
$device-property_set('BLOCK_SIZE', $opt_blocksize)
my $block_size = $device-property_get(block_size);
my $blocksize = $device-property_get(BLOCK_SIZE);
$blocksize_line = # add device_property \BLOCK_SIZE\ \$blocksize\
to the device;
It should be upper case right.
$ amtapetype -o length=400g -b 256k -f /dev/rmt/0n
Applying heuristic check for compression.
blocksize = 262144 blocks written = 1770
blocksize = 262144 blocks written = 1770
Wrote random (uncompressible) data at 7364998.0952381 bytes/sec
Wrote fixed (compressible) data at 38666240 bytes/sec
Compression: enabled
Writing one file to fill the volume.
blocksize = blocks written = 1552513
amtapetype: Wrote less than 1MB to the device: Error writing block: Mysterious
short write on tape device: Tried 262144, got 0
-- GG
-Ursprungligt meddelande-
Från: owner-amanda-us...@amanda.org [mailto:owner-amanda-us...@amanda.org] För
Gunnarsson, Gunnar
Skickat: den 20 maj 2010 21:24
Till: 'Dustin J. Mitchell'
Kopia: amanda-users@amanda.org
Ämne: SV: amtapetype Amanda 3.1.0beta2
Yes it ran for a quite long time.
I added a print in the script see below. It will run over the night.
-- GG
amtapetype -o length=400g -b 256k -f /dev/rmt/0n Applying heuristic check for
compression.
blocksize = 262144 blocks written = 1770 blocksize = 262144 blocks written =
1770 Wrote random (uncompressible) data at 7364998.0952381 bytes/sec Wrote
fixed (compressible) data at 38666240 bytes/sec
Compression: enabled
Writing one file to fill the volume.
-Ursprungligt meddelande-
Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell
Skickat: den 20 maj 2010 17:44
Till: Gunnarsson, Gunnar
Kopia: amanda-users@amanda.org
Ämne: Re: amtapetype Amanda 3.1.0beta2
On Thu, May 20, 2010 at 3:00 AM, Gunnarsson, Gunnar gunnar.gunnars...@svk.se
wrote:
Applying heuristic check for compression.
Wrote random (uncompressible) data at 7492244.64516129 bytes/sec Wrote
fixed (compressible) data at 35732243.6923077 bytes/sec
Compression: enabled
Writing one file to fill the volume.
File 1, block 6910315
amtapetype: Wrote less than 1MB to the device: Error writing block:
Mysterious short write on tape device: Tried 32768, got 0
That's odd.. it looks like it wrote 6910315 blocks, yet when it multiplied that
by the block size, it got something less than 1MB.
Did this take a long time, or happen immediately? Can you add some print's
around line 186 of amtapetype to print the blocksize and number of blocks?
Dustin
--
Open Source Storage Engineer
http://www.zmanda.com
SV: amtapetype Amanda 3.1.0beta2
Hi, Still problem see below. I'm running Solaris 10 sparc witch HP Ultrium LTO 3 drive ( SL48 Tape Library ). Thanks Gunnar Applying heuristic check for compression. Wrote random (uncompressible) data at 7492244.64516129 bytes/sec Wrote fixed (compressible) data at 35732243.6923077 bytes/sec Compression: enabled Writing one file to fill the volume. File 1, block 6910315 amtapetype: Wrote less than 1MB to the device: Error writing block: Mysterious short write on tape device: Tried 32768, got 0 -Ursprungligt meddelande- Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell Skickat: den 18 maj 2010 17:01 Till: Gunnarsson, Gunnar Kopia: amanda-users@amanda.org Ämne: Re: amtapetype Amanda 3.1.0beta2 On Tue, May 18, 2010 at 9:37 AM, Dustin J. Mitchell dus...@zmanda.com wrote: Hmm.. can you send me the syscall trace (truss or strace) of that? (truss now in hand..) Oh, silly me - the code detects that the tape drive needs the FSF_AFTER_FILEMARK property, by virtue of getting an error from the device .. but then fails on that very error. This should be pretty easy to fix - try this patch: http://github.com/djmitche/amanda/commit/fsf_after_filemark.patch You can probably apply that in-place without too much trouble. Dustin -- Open Source Storage Engineer http://www.zmanda.com amtapetype.20100519082823.debug Description: amtapetype.20100519082823.debug
HW Compression
Hi, On the Amanada Wiki there is a note on Solaris: Set ST_MODE_SEL_COMP in st.conf for your device and then use the correct tape device variant (do not use c and u device) How can I verify that it is being set - the driver is seen and configured by the system. I want to use the driver in a none compression mode. # mt -f /dev/rmt/0n status HP Ultrium LTO 3 tape drive: sense key(0x0)= No Additional Sense residual= 0 retries= 0 file no= 2 block no= 0 The drive is sitting in a SL48 Tape library and the host is Solaris Sparc runnint Solaris 10. Sorry this is more of an os question but still I give it a try. Thanks Gunnar
SV: amtapetype Amanda 3.1.0beta2
Yes it ran for a quite long time. I added a print in the script see below. It will run over the night. -- GG amtapetype -o length=400g -b 256k -f /dev/rmt/0n Applying heuristic check for compression. blocksize = 262144 blocks written = 1770 blocksize = 262144 blocks written = 1770 Wrote random (uncompressible) data at 7364998.0952381 bytes/sec Wrote fixed (compressible) data at 38666240 bytes/sec Compression: enabled Writing one file to fill the volume. -Ursprungligt meddelande- Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell Skickat: den 20 maj 2010 17:44 Till: Gunnarsson, Gunnar Kopia: amanda-users@amanda.org Ämne: Re: amtapetype Amanda 3.1.0beta2 On Thu, May 20, 2010 at 3:00 AM, Gunnarsson, Gunnar gunnar.gunnars...@svk.se wrote: Applying heuristic check for compression. Wrote random (uncompressible) data at 7492244.64516129 bytes/sec Wrote fixed (compressible) data at 35732243.6923077 bytes/sec Compression: enabled Writing one file to fill the volume. File 1, block 6910315 amtapetype: Wrote less than 1MB to the device: Error writing block: Mysterious short write on tape device: Tried 32768, got 0 That's odd.. it looks like it wrote 6910315 blocks, yet when it multiplied that by the block size, it got something less than 1MB. Did this take a long time, or happen immediately? Can you add some print's around line 186 of amtapetype to print the blocksize and number of blocks? Dustin -- Open Source Storage Engineer http://www.zmanda.com
SV: amtapetype Amanda 3.1.0beta2
Thanks Dustin the patch is working fine, good work. Compression enabled, does it mean that it is using compression ? -- GG bash-3.00$ amtapetype -f arkive /dev/rmt/0n Applying heuristic check for compression. Wrote random (uncompressible) data at 7492244.64516129 bytes/sec Wrote fixed (compressible) data at 35732243.6923077 bytes/sec Compression: enabled Writing one file to fill the volume. -Ursprungligt meddelande- Från: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell Skickat: den 18 maj 2010 17:01 Till: Gunnarsson, Gunnar Kopia: amanda-users@amanda.org Ämne: Re: amtapetype Amanda 3.1.0beta2 On Tue, May 18, 2010 at 9:37 AM, Dustin J. Mitchell dus...@zmanda.com wrote: Hmm.. can you send me the syscall trace (truss or strace) of that? (truss now in hand..) Oh, silly me - the code detects that the tape drive needs the FSF_AFTER_FILEMARK property, by virtue of getting an error from the device .. but then fails on that very error. This should be pretty easy to fix - try this patch: http://github.com/djmitche/amanda/commit/fsf_after_filemark.patch You can probably apply that in-place without too much trouble. Dustin -- Open Source Storage Engineer http://www.zmanda.com
amtapetype Amanda 3.1.0beta2
Hi, I'm testing the new chg-robot interface and it seems to be working. However i got som problem with amtapetype se debug below. Tue May 18 13:21:50 2010: amtapetype: Building type TAPESTART header of 32768-32768 bytes with name='TEST-001' disk='' dumplevel=0 and blocksize=32768 Tue May 18 13:21:53 2010: amtapetype: Building type FILE header of 32768-32768 bytes with name='localhost' disk='/test1' dumplevel=0 and blocksize=32768 Tue May 18 13:21:56 2010: amtapetype: Building type FILE header of 32768-32768 bytes with name='localhost' disk='/test2' dumplevel=0 and blocksize=32768 Tue May 18 13:21:59 2010: amtapetype: Building type FILE header of 32768-32768 bytes with name='localhost' disk='/test3' dumplevel=0 and blocksize=32768 Tue May 18 13:22:08 2010: amtapetype: Starting x...@5245d8 (xfersourcedev...@523c08 - xferdestn...@52f018) Tue May 18 13:22:08 2010: amtapetype: Final linkage: xfersourcedev...@523c08 -(PULL_BUFFER)- xferelementg...@530800 -(PUSH_BUFFER)- xferdestn...@52f018 Tue May 18 13:22:08 2010: amtapetype: Device /dev/rmt/0n error = 'EOF' Tue May 18 13:22:08 2010: amtapetype: Device /dev/rmt/0n error = 'Could not seek to file 2 correctly; got 1' Tue May 18 13:22:08 2010: amtapetype: Device /dev/rmt/0n setting status flag(s): DEVICE_STATUS_DEVICE_ERROR Tue May 18 13:22:08 2010: amtapetype: critical (fatal): Error setting FSF_AFTER_FILEMARK: Could not seek to file 2 correctly; got 1 at /opt/sfw/sbin/amtapetype line 389. Thanks Gunnar
SV: runtar error that I do not understand
Sun is delivering old versions of GNU tar: Written by John Gilmore and Jay Fenlason. bash-3.00$ /usr/sfw/bin/gtar --version tar (GNU tar) 1.14 Copyright (C) 2004 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. You need version with --no-check-device in order to backup snapshot in zfs. So get the latest version of SMCtar and it should work fine. bash-3.00$ /usr/local/bin/tar --version tar (GNU tar) 1.20 Copyright (C) 2008 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Best regards GG Dustin J. Mitchell Sent: Monday, May 03, 2010 5:42 PM To: McGraw, Robert P Cc: amanda-users@amanda.org Subject: Re: runtar error that I do not understand On Mon, May 3, 2010 at 4:31 PM, McGraw, Robert P rmcg...@purdue.edu wrote: If I manually run the following gtar command I get a segmentation fault /usr/sfw/bin/gtar --create Uh-oh .. time to open a Sun^WOracle support case? It looks like a bug either in ZFS or GNU Tar.. Dustin -- Open Source Storage Engineer http://www.zmanda.com [McGraw, Robert P] Have you every tried star. I though I would give it a shot and see if it work ok at least until I can get the GNU tar working. Thanks Robert
SV: SV: runtar error that I do not understand
Yes for Solaris 10 but we are still running Solaris 8 so we choose to use SMCtar packages instead. Solaris 10 10/09 has Gnu tar version 1.23 Till: Gunnarsson, Gunnar Kopia: 'McGraw, Robert P'; 'Dustin J. Mitchell'; 'amanda-users@amanda.org' Ämne: Re: SV: runtar error that I do not understand That all depends on whether you are up on your patches. I went and checked my Solaris 10 system and found: # /usr/sfw/bin/gtar --version tar (GNU tar) 1.22 Copyright (C) 2009 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html. This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Written by John Gilmore and Jay Fenlason. That is the gnutar release from 3/09, and the only newer one is 1.23 from 3/10. But, I had done an upgrade late last year (using liveupdate and the 5/09 release), and have also applied the latest recommended and security patch bundles within the last few months. So, I have # more /etc/release Solaris 10 5/09 s10s_u7wos_08 SPARC Copyright 2009 Sun Microsystems, Inc. All Rights Reserved. Use is subject to license terms. Assembled 30 March 2009 All of which means, check what your system actually has (and keep it up to date). --- Chris Hoogendyk - O__ Systems Administrator c/ /'_ --- Biology Geology Departments (*) \(*) -- 140 Morrill Science Center ~~ - University of Massachusetts, Amherst hoogen...@bio.umass.edu --- Erdös 4 Gunnarsson, Gunnar wrote: Sun is delivering old versions of GNU tar: Written by John Gilmore and Jay Fenlason. bash-3.00$ /usr/sfw/bin/gtar --version tar (GNU tar) 1.14 Copyright (C) 2004 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. You need version with --no-check-device in order to backup snapshot in zfs. So get the latest version of SMCtar and it should work fine. bash-3.00$ /usr/local/bin/tar --version tar (GNU tar) 1.20 Copyright (C) 2008 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Best regards GG Dustin J. Mitchell Sent: Monday, May 03, 2010 5:42 PM To: McGraw, Robert P Cc: amanda-users@amanda.org Subject: Re: runtar error that I do not understand On Mon, May 3, 2010 at 4:31 PM, McGraw, Robert P rmcg...@purdue.edu wrote: If I manually run the following gtar command I get a segmentation fault /usr/sfw/bin/gtar --create Uh-oh .. time to open a Sun^WOracle support case? It looks like a bug either in ZFS or GNU Tar.. Dustin -- Open Source Storage Engineer http://www.zmanda.com [McGraw, Robert P] Have you every tried star. I though I would give it a shot and see if it work ok at least until I can get the GNU tar working. Thanks Robert --
SV: FAILED [spawn /bin/gzip: dup2 out: Bad file descriptor] on 2.6.1p2
Switching to tcp instead of using udp cured those problems. -- GG -Ursprungligt meddelande- Från: owner-amanda-us...@amanda.org [mailto:owner-amanda-us...@amanda.org] För Dustin J. Mitchell Skickat: den 13 april 2010 18:11 Till: Volker Pallas Kopia: amanda-users@amanda.org Ämne: Re: FAILED [spawn /bin/gzip: dup2 out: Bad file descriptor] on 2.6.1p2 On Mon, Apr 12, 2010 at 4:48 AM, Volker Pallas ama...@sqmail.de wrote: Hi, I'm having a bit of a problem on *some* servers concerning failed backups with the error message: lev # FAILED [spawn /bin/gzip: dup2 out: Bad file descriptor] Gunnar had a similar problem - maybe his experience will help? http://www.mail-archive.com/amanda-users@amanda.org/msg42119.html Dustin -- Open Source Storage Engineer http://www.zmanda.com
SV: Build problem Solaris 10 Amanda 2.6.1p2
I found an explanation, old version of Amanda was lying around on the build machine and the installation process picked the symbol from there and added to the librestore library. After having cleaned up the system the build process works using gcc that is bundled with the system. Thanks for your help. GG On Tue, Jan 19, 2010 at 4:32 AM, Gunnarsson, Gunnar gunnar.gunnars...@svk.se wrote: amrestore and amfetchdump don't build correctly on my system - # /opt/sfw/sbin/amrestore ld.so.1: amrestore: fatal: relocation error: file /opt/sfw/lib/amanda/librestore-2.6.1p2.so: symbol getconf_str: referenced symbol not found Killed #/opt/sfw/sbin/amfetchdump ld.so.1: amfetchdump: fatal: relocation error: file /opt/sfw/lib/amanda/librestore-2.6.1p2.so: symbol getconf_str: referenced symbol not found Other binaries seems to be working. Odd! getconf_str is a macro defined in conffile.h, so the linker should never see it. The reference is most likely from restore-src/restore.c, which doesn't actually include conffile.h. However, find.h includes diskfile.h which includes conffile.h, so the macro definition should be available all the same.
Build problem Solaris 10 Amanda 2.6.1p2
Hi, amrestore and amfetchdump don't build correctly on my system - # /opt/sfw/sbin/amrestore ld.so.1: amrestore: fatal: relocation error: file /opt/sfw/lib/amanda/librestore-2.6.1p2.so: symbol getconf_str: referenced symbol not found Killed #/opt/sfw/sbin/amfetchdump ld.so.1: amfetchdump: fatal: relocation error: file /opt/sfw/lib/amanda/librestore-2.6.1p2.so: symbol getconf_str: referenced symbol not found Other binaries seems to be working. My build environment is Solaris 10 Sparc and I'm using standard gcc tool included on the system: CFLAGS: -g -O2 -fno-strict-aliasing -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 CPPFLAGS: -D_REENTRANT -I/opt/sfw/include -fno-strict-aliasing -D_GNU_SOURCE -D_REENTRANT -D_PTHREADS -I/usr/local/include/glib-2.0 -I/usr/local/lib/glib-2.0/include LDFLAGS: -L/opt/sfw/lib -R/opt/sfw/lib -L/opt/sfw/lib LIBS: -lm -L/usr/local/lib -lgmodule-2.0 -lgobject-2.0 -lgthread-2.0 -lpthread -lthread -lrt -lglib-2.0 -liconv -lsocket -lnsl -lresolv -lintl -lcrypto -L/opt/sfw/lib -lcurl -lz -lssl -lcrypto -lsocket -lnsl -lz -lnsl -lsocket Amanda Components: server restore (no client) (no amrecover) Working IPv6: no Build documentation: no Directories: Application: /opt/sfw/libexec/amanda/application Configuration: /etc/opt/amanda GNU Tar lists: /var/opt/lib/amanda/gnutar-lists Perl modules (amperldir): /usr/perl5/site_perl/5.8.4 Template and example datpwda files (amdatadir): /opt/sfw/share/amanda Temporary: /tmp/amanda Thanks Gunnar Gunnarsson
SV: Amanda Windows client 2.6.1 released
I have been running the 2.6 client and the server is complaing about estimate failures see below: planner: disk svkstas2:E:/, estimate of level 2 failed. planner: disk svkstas2:C:/, estimate of level 2 failed. planner: disk svkptas1:E:/, estimate of level 2 failed. planner: disk svkptas1:C:/, estimate of level 2 failed. planner: disk svkstas1:E:/, estimate of level 2 failed. planner: disk svkstas1:C:/, estimate of level 2 failed. Is this a know problem and solved in the 2.6.1 version ? -- Thanks Gunnar Gunnarsson Amanda Windows Client 2.6.1 has been released. This release has been tested with Amanda 2.6.1 server. New features (compared to Amanda windows client 2.6) - 32bit and 64bit client are available - Support for Windows 7 and Windows 2008 server R2 - Installshield based installer - Support for exclude patterns (wildcard * is supported) - ZWCconfig tool for configuring client parameters (port numbers, Amanda server name)
SV: SL48 autoloader
Yes Chris you are right it's an OS issue and I've opened a case with SUN support. We want to use the sgen - Generic SCSI device driver to access the library in Solaris and the tape drivers. And then hopefully the mtx changer support in Amanda. Thanks Gunnar Gunnarsson Gunnarsson, Gunnar wrote: Hi I'm trying to configure Sun SL48 autloader with Amanda on Solaris 10 10/09 sparc system. SL48 library is connected to our SAN switches and disk are beeing used and configured on the SAN but I'm not seeing the the tape driver. # ls /dev/rmt/ # # /usr/sbin/add_drv -v st exit status = 0 devfsadm: driver failed to attach: st exit status = 1 Warning: Driver (st) successfully added to system but failed to attach Driver (st) installed. Your post is really an OS question rather than an Amanda question. I see that you also posted on the Sun forums, so that's good. I also found a post (I just googled sl48 san solaris) by someone else that has a fair number of replies that might help. It's someone trying to do the same thing, but with NetBackup (which is really irrelevant). The clues appear to be related to configuring sg.conf and st.conf. That makes sense to me, but I've never messed with devices connected through SAN switches. The thread is here http://www.backupcentral.com/phpBB2/two-way-mirrors-of-external-mailing-lists-3/symantec-netbackup-18/solaris-10-run-netbackup-cannot-detect-sun-storageteksl48-96060/ -- --- Chris Hoogendyk - O__ Systems Administrator c/ /'_ --- Biology Geology Departments (*) \(*) -- 140 Morrill Science Center ~~ - University of Massachusetts, Amherst hoogen...@bio.umass.edu --- Erdös 4
SL48 autoloader
Hi I'm trying to configure Sun SL48 autloader with Amanda on Solaris 10 10/09 sparc system. SL48 library is connected to our SAN switches and disk are beeing used and configured on the SAN but I'm not seeing the the tape driver. # ls /dev/rmt/ # # /usr/sbin/add_drv -v st exit status = 0 devfsadm: driver failed to attach: st exit status = 1 Warning: Driver (st) successfully added to system but failed to attach Driver (st) installed. Thanks Gunnar Gunnarsson
SV: is amanda the right solution
Its capabilities will be expanded -- it will be able to re-assemble and re-split dumps, so the source and destination volume sizes need not match. The command will still be separate from the usual amdump runs, and won't do the usual report-sending or log-rotating. To put it another way, amvault will benefit from the improvements being made to how changers, devices and data are handled, but I won't have a chance to address the higher-level questions of scheduling, metadata maintenance, tertiary tape cycles, etc. Good working Dustin. One more question on amvault what would be the result of calling 'amvault latest' for a giving configuration - I would expect to get the most resent 0 dump on each volume in the disk list is that correct ? -- GG
fssnap application
Hi, Vi have busy file system in Solaris 8 were ufsdump/gtar doesn't work so well. Instead we use fssnap (ufs) capabilities with tar. It would be nice to implement it as an application similar to zfs snapshot ! -- GG
SV: is amanda the right solution
Yes this is what we need. What are the plans for amvault in the upcoming releases of Amanda ? Do you have any how Tos on usage. From: djmit...@gmail.com [mailto:djmit...@gmail.com] För Dustin J. Mitchell On Fri, Jul 10, 2009 at 10:03 AM, Gunnarsson, Gunnargunnar.gunnars...@svk.se wrote: We have a need to do some selective archiving e.g. copies of selective 0 dumps with or without indexes. amvault is designed for this sort of thing, although in 2.6.1 it's experimental and somewhat limited in its capabilities.
SV: filesys for tape?
We are using zfs filesystem running Solaris on Sparc HW and it's working well, x64 HW is an option for Solaris as well. Go for it. -- GG What filesystem and mount options are recommended for a tapetype harddisk? Details: The amanda-server is a Debian 5 (lenny) host. The tape size is 40 gigs. The filesystems are about 540, 990, and 950 gigs.
SV: is amanda the right solution
Are there any applications adds on being developed ? We have a need to do some selective archiving e.g. copies of selective 0 dumps with or without indexes. -- GG -Ursprungligt meddelande- Från: owner-amanda-us...@amanda.org [mailto:owner-amanda-us...@amanda.org] För Dustin J. Mitchell Skickat: den 7 juli 2009 22:03 Till: Andreas Kuntzagk Kopia: amanda-users@amanda.org Ämne: Re: is amanda the right solution On Wed, Jul 1, 2009 at 9:21 AM, Andreas Kuntzagkandreas.kuntz...@mdc-berlin.de wrote: we have some genome sequencers which produce lots of data. ATM these are stored on 2 big Solaris Fileservers (~34TB each) but in the long term we want to move these to some kind of archive. One idea is to bye a second tape library (we already have one SL24) and write these datasets to tapes. Since these datasets can span more then one tape (upto 3.5 TB) just using tar or dump is not the solution (at least I'm not aware how to do this). So can I use amanda for this? I found something about long-term archiving in the faq but it is not to elaborate. I'm surprised nobody wanted to answer this. Amanda is really designed around the idea of repeatedly backing up the same data, rather than doing one-off dumps. Folks certainly have cajoled Amanda into doing what you describe, through manual configuration (editing the disklist all the time). If you're feeling more motivated, we're to the point where Amanda has the tools to do this sort of thing for you -- with a bit of Perl and some studying of the Amanda::* POD, you could write an application that takes a stream as input (say, a tarball) and write it to tape: tar -czf - /my/sequences | write-to-tapes SequenceBackupou working and still use the usual Amanda recovery tools to read that data (amfetchdump, probably -- amrecover would be useless since there are no indices, but I doubt you need indices). If you're interested in working on a script like this, I'll be happy to help. Dustin -- Open Source Storage Engineer http://www.zmanda.com
sendbackup: critical (fatal): Can't dup2: Bad file number
Se below debug listing. Is this a know issue with dup2 ? Thans Gunnar Gunnarsson 1241554091.665921: sendbackup: pid 20149 ruid 91 euid 91 version 2.6.1: start at Tue May 5 22:08:11 2009 1241554091.666358: sendbackup: Version 2.6.1 1241554091.668434: sendbackup: pid 20149 ruid 91 euid 91 version 2.6.1: rename at Tue May 5 22:08:11 2009 1241554091.670599: sendbackup: Parsed request as: program `amgtar' 1241554091.670627: sendbackup: disk `/' 1241554091.670648: sendbackup: device `/' 1241554091.670669: sendbackup: level 2 1241554091.670690: sendbackup: since NODATE 1241554091.670710: sendbackup: options `' 1241554091.671351: sendbackup: pipespawnv: stdoutfd is 0 1241554091.672033: sendbackup: Spawning /opt/sfw/libexec/amanda/application/amzfs-snapshot amzfs-snapshot PRE-DLE-BACKUP --execute-where client --config hansa --host hansabck --disk / --device / --level 2 --pfexec YES --pfexec-path /usr/bin/pfexec in pipeline 1241554093.912552: sendbackup: script: PROPERTY directory //.zfs/snapshot/amanda-_-current 1241554093.924144: sendbackup: pipespawnv: stdoutfd is 50 1241554093.924271: sendbackup: Spawning /usr/bin/gzip /usr/bin/gzip --fast in pipeline 1241554093.925843: sendbackup: gnutar: pid 20165: /usr/bin/gzip1241554093.925981: sendbackup: pid 20165: /usr/bin/gzip --fast 1241554093.926548: sendbackup: pipespawnv: stdoutfd is 5120 1241554093.926930: sendbackup: Spawning /opt/sfw/libexec/amanda/application/amgtar amgtar support --config hansa --host hansabck --disk / --device / in pipeline 1241554093.996508: sendbackup: support line: CONFIG YES 1241554093.996628: sendbackup: support line: HOST YES 1241554093.996658: sendbackup: support line: DISK YES 1241554093.996686: sendbackup: support line: MAX-LEVEL 9 1241554093.996735: sendbackup: support line: INDEX-LINE YES 1241554093.996765: sendbackup: support line: INDEX-XML NO 1241554093.996793: sendbackup: support line: MESSAGE-LINE YES 1241554093.996820: sendbackup: support line: MESSAGE-XML NO 1241554093.996847: sendbackup: support line: RECORD YES 1241554093.996875: sendbackup: support line: INCLUDE-FILE YES 1241554093.996903: sendbackup: support line: INCLUDE-LIST YES 1241554093.996932: sendbackup: support line: INCLUDE-OPTIONAL YES 1241554093.996960: sendbackup: support line: EXCLUDE-FILE YES 1241554093.996989: sendbackup: support line: EXCLUDE-LIST YES 1241554093.997018: sendbackup: support line: EXCLUDE-OPTIONAL YES 1241554093.997046: sendbackup: support line: COLLECTION NO 1241554093.997075: sendbackup: support line: MULTI-ESTIMATE YES 1241554093.997104: sendbackup: support line: CALCSIZE YES 1241554094.002144: sendbackup: sendbackup: running /opt/sfw/libexec/amanda/application/amgtar 1241554094.002274: sendbackup: backup 1241554094.002297: sendbackup: --message 1241554094.002317: sendbackup: line 1241554094.002337: sendbackup: --config 1241554094.002356: sendbackup: hansa 1241554094.002376: sendbackup: --host 1241554094.002396: sendbackup: hansabck 1241554094.002416: sendbackup: --disk 1241554094.002435: sendbackup: / 1241554094.002454: sendbackup: --device 1241554094.002473: sendbackup: / 1241554094.002493: sendbackup: --level 1241554094.002512: sendbackup: 2 1241554094.002531: sendbackup: --index 1241554094.002550: sendbackup: line 1241554094.002569: sendbackup: --record 1241554094.002589: sendbackup: --atime-preserve 1241554094.002608: sendbackup: NO 1241554094.002627: sendbackup: --check-device 1241554094.002647: sendbackup: NO 1241554094.002666: sendbackup: --directory 1241554094.002686: sendbackup: //.zfs/snapshot/amanda-_-current 1241554094.002706: sendbackup: 1241554094.003034: sendbackup: critical (fatal): Can't dup2: Bad file number 1241554094.003991: sendbackup: error: sendbackup: Can't dup2: Bad file number 1241554094.043001: sendbackup: pipespawnv: stdoutfd is 171096 1241554094.043703: sendbackup: Spawning /opt/sfw/libexec/amanda/application/amzfs-snapshot amzfs-snapshot POST-DLE-BACKUP --execute-where client --config hansa --host hansabck --disk / --device / --level 2 --pfexec YES --pfexec-path /usr/bin/pfexec in pipeline 1241554095.704958: sendbackup: pid 20149 finish time Tue May 5 22:08:15 2009
SV: SUN SL24 autoload
We haven't bought it yet, I thought I might ask on the list. I've used stctl but that driver is not beeing maintained any more. -- Gunnar -Ursprungligt meddelande- Från: Chris Hoogendyk [mailto:hoogen...@bio.umass.edu] Skickat: den 17 april 2009 17:57 Till: Gunnarsson, Gunnar Kopia: amanda-users@amanda.org Ämne: Re: SUN SL24 autoload Gunnarsson, Gunnar wrote: Anyone using Sun SL24 autoloader in amanda configuration on Sun Sparc platform ? Can you tell us more? I don't have the SL24, but, if it works for Solaris, it ought to just work for Amanda (via mt and mtx). I did find an exchange on the netbackup list (on backupcentral) where they were having issues with this setup, and someone pointed out a Sun patch issue that affected the drivers: http://www.auscert.org.au/render.html?it=9490 (linked to from http://www.backupcentral.com/phpBB2/two-way-mirrors-of-external-mailing-lists-3/symantec-netbackup-18/mpt-driver-and-sg-91205/) Don't know if that will solve your problem, but, with more information, someone on the list may be able to help. I don't see it on the mtx compatibility list (http://mtx.opensource-sw.net/compatibility.php?sorttype=1start=1count=0); but, then, that is a user contributed list, and people may just not take the time to contribute. I'm going to add my own library now, because it isn't on there either (and I've been using it successfully for quite some time now). -- --- Chris Hoogendyk - O__ Systems Administrator c/ /'_ --- Biology Geology Departments (*) \(*) -- 140 Morrill Science Center ~~ - University of Massachusetts, Amherst hoogen...@bio.umass.edu --- Erdös 4
SUN SL24 autoload
Anyone using Sun SL24 autoloader in amanda configuration on Sun Sparc platform ? Thanks Gunnar Gunnarsson
Amanda zfs
I'm been using amzfs-snapshot and it works well. One feature that I would like to have in the plug-in is an option to run a script before and after a snapshot is created thus I can shutdown my application create a snapshot bring it online and proceed with the backup. Why not add a general script application that could be user for rman backup for oracle as well in the distribution. Thanks Gunnar Gunnarsson
SV: Problems with chg-manual changer script - Amanda 2.6.0p2
I had to change another line in the chg-manual script. I don't no if his is a mt specific problem in Solaris but hera are my changes for chg-lib.sh diff /opt/sfw/libexec/amanda/chg-lib.sh /opt/sfw/libexec/amanda/chg-lib.sh.off if echo $amdevcheck_output | $EGREP VOLUME_MISSING|VOLUME_ERROR /dev/null; then --- if echo $amdevcheck_output | $EGREP VOLUME_MISSING /dev/null; then 134c134 $MT $MTF $try_eject_device_tape rewoffl --- $MT $MTF $try_eject_device_tape eject Thanks Gunnar Gunnarsson -Ursprungligt meddelande- Från: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] För Jon LaBadie Skickat: den 25 oktober 2008 17:02 Till: amanda-users@amanda.org Ämne: Re: Problems with chg-manual changer script - Amanda 2.6.0p2 On Sat, Oct 25, 2008 at 02:24:47PM +0200, Gunnarsson, Gunnar wrote: Hey everyone, I'm having problem setting up chg-manual changer script in Solaris 10. From the taped.debug: mt: unknown command: eject I tried to change the eject command to rewoffl instead # diff /opt/sfw/libexec/amanda/chg-lib.sh /opt/sfw/libexec/amanda/chg-lib.sh.off 134c134 $MT $MTF $try_eject_device_tape rewoffl --- $MT $MTF $try_eject_device_tape eject But then I got some other error instead see below. Any ideas what might be the problem ? 1224898398.489880: taper: putresult: 26 REQUEST-NEW-TAPE 1224898398.490892: taper: getcmd: NEW-TAPE 1224898405.934941: taper: changer_query: changer return was 5 1 1224898405.935141: taper: changer_query: searchable = 0 1224898405.935176: taper: changer_find: looking for NULL changer is searchable = 0 Error rewinding device /dev/rmt/1n Error rewinding device /dev/rmt/1n 1224898466.433921: taper: device_read_label; mode = 0 Can't open tape device /dev/rmt/1n: Device busy /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy Just a WAG. Only thing in common is Solaris, but it was 9 vs 10. My mt command often would return while the tape was still rewinding. If run from the command line, I would get the next shell prompt and yet could still hear the tape drive and see the tape's activity lights. Until the tape had actually completed its task, new commands would fail. In my scripts, I had to insert sleep X (X to be empirically defined) after mt cmds that might take a significant time. jl -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 12027 Creekbend Drive (703) 787-0884 Reston, VA 20194 (703) 787-0922 (fax)
SV: Problems with chg-manual changer script - Amanda 2.6.0p2
But I'm using the chg-manual changer script , it is configured to send email when next tape should be loaded. $ cat changer.conf # # Changer config file (changerfile) # resend_mail=36000 # 10 timmar timeout_mail=604800 # 7 days request=email amanda.conf tpchanger chg-manual # the tape-changer glue script changerfile /etc/opt/amanda/hansa-arkive/changer runtapes 4 How are the labels written when using more than one tape ? Thanks Gunnar Gunnarsson -Ursprungligt meddelande- Från: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] För Dustin J. Mitchell Skickat: den 26 oktober 2008 20:19 Till: Gunnarsson, Gunnar Kopia: amanda-users@amanda.org Ämne: Re: Problems with chg-manual changer script - Amanda 2.6.0p2 On Sun, Oct 26, 2008 at 12:40 PM, Gunnarsson, Gunnar [EMAIL PROTECTED] wrote: I had only written label on one tape before I ran amdump, could that explain my problems below, runtapes is set to 4 in my config. No, you're definitely seeing weird behavior from your tape changer. You may want to increase initial_poll_delay in your changer.conf so that it gives the tape drive more time to unload while scanning for tapes. Dustin -- Storage Software Engineer http://www.zmanda.com
SV: Problems with chg-manual changer script - Amanda 2.6.0p2
I had only written label on one tape before I ran amdump, could that explain my problems below, runtapes is set to 4 in my config. On Sat, Oct 25, 2008 at 02:24:47PM +0200, Gunnarsson, Gunnar wrote: Hey everyone, I'm having problem setting up chg-manual changer script in Solaris 10. From the taped.debug: mt: unknown command: eject I tried to change the eject command to rewoffl instead # diff /opt/sfw/libexec/amanda/chg-lib.sh /opt/sfw/libexec/amanda/chg-lib.sh.off 134c134 $MT $MTF $try_eject_device_tape rewoffl --- $MT $MTF $try_eject_device_tape eject But then I got some other error instead see below. Any ideas what might be the problem ? 1224898398.489880: taper: putresult: 26 REQUEST-NEW-TAPE 1224898398.490892: taper: getcmd: NEW-TAPE 1224898405.934941: taper: changer_query: changer return was 5 1 1224898405.935141: taper: changer_query: searchable = 0 1224898405.935176: taper: changer_find: looking for NULL changer is searchable = 0 Error rewinding device /dev/rmt/1n Error rewinding device /dev/rmt/1n 1224898466.433921: taper: device_read_label; mode = 0 Can't open tape device /dev/rmt/1n: Device busy /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy Just a WAG. Only thing in common is Solaris, but it was 9 vs 10. My mt command often would return while the tape was still rewinding. If run from the command line, I would get the next shell prompt and yet could still hear the tape drive and see the tape's activity lights. Until the tape had actually completed its task, new commands would fail. In my scripts, I had to insert sleep X (X to be empirically defined) after mt cmds that might take a significant time. jl -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 12027 Creekbend Drive (703) 787-0884 Reston, VA 20194 (703) 787-0922 (fax)
Problems with chg-manual changer script - Amanda 2.6.0p2
Hey everyone, I'm having problem setting up chg-manual changer script in Solaris 10. From the taped.debug: mt: unknown command: eject I tried to change the eject command to rewoffl instead # diff /opt/sfw/libexec/amanda/chg-lib.sh /opt/sfw/libexec/amanda/chg-lib.sh.off 134c134 $MT $MTF $try_eject_device_tape rewoffl --- $MT $MTF $try_eject_device_tape eject But then I got some other error instead see below. Any ideas what might be the problem ? 1224898398.489880: taper: putresult: 26 REQUEST-NEW-TAPE 1224898398.490892: taper: getcmd: NEW-TAPE 1224898405.934941: taper: changer_query: changer return was 5 1 1224898405.935141: taper: changer_query: searchable = 0 1224898405.935176: taper: changer_find: looking for NULL changer is searchable = 0 Error rewinding device /dev/rmt/1n Error rewinding device /dev/rmt/1n 1224898466.433921: taper: device_read_label; mode = 0 Can't open tape device /dev/rmt/1n: Device busy /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy /dev/rmt/1n: Device busy Can't open tape device /dev/rmt/1n: Device busy Thanks Gunnar Gunnarsson
chg-manual changer-script Amanda 2.6.0p2
Hi, I noticed that chg-manual was doing eject with mt command on Solaris 10 which is a unknown command see below. Instead offline, rewoffl should be used I assume. From the taper.debug: mt: unknown command: eject Thanks GG
SV: Backup using ZFS snapshots on OpenSolaris
Amanda backup user needs to run zfs snapshot/destroy, zfs allow -ldu AMANDA_USER mount,snapshot,destroy FILE SYSTEM is still a missing feature in any Solaris 10 Update that I know of. Workaround ? thanks GG -Ursprungligt meddelande- Från: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] För Nick Smith Skickat: den 1 oktober 2008 16:42 Till: amanda-users@amanda.org Ämne: Re: Backup using ZFS snapshots on OpenSolaris Be aware that there isn't an offical beta (i.e. tagged versions) as yet - it's a snapshot of the Amanda 2.6.1 *alpha* *development* SVN tree. Having said that Amanda built from SVN HEAD works pretty well! Q : Is anyone planning to backup ZFS volumes with Amanda? Regards, Nick Smith Martin.Hepworth wrote: Gunnar Latest beta has this support -- Martin Hepworth Snr Systems Administrator Solid State Logic Tel: +44 (0)1865 842300 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Gunnarsson, Gunnar Sent: 01 October 2008 14:42 To: amanda-users@amanda.org Subject: Backup using ZFS snapshots on OpenSolaris Hi all ZFS users, I've been planning to upgrade to version 2.6.0 and start using the script-tools mechanism in amanda for backing up ZFS with snapshot. Are there any open source plug-in available for doing this ? thanks GG
Backup using ZFS snapshots on OpenSolaris
Hi all ZFS users, I've been planning to upgrade to version 2.6.0 and start using the script-tools mechanism in amanda for backing up ZFS with snapshot. Are there any open source plug-in available for doing this ? thanks GG
SV: Backups to disk and tape
On Sat, Aug 9, 2008 at 9:22 AM, Peter Spikings [EMAIL PROTECTED] wrote: Is that possible? (because I can't see how from reading the man pages). What you're asking about is called D2D2T, and Amanda doesn't support it directly just yet, although preliminary support is a goal for the next release or so. RAIT is the current best option. As long as you're mirroring, there's no problem with deleting a vtape after 7 days. If you need to recover that data, you'll just retrieve the tape and use it directly. So a little bit of shell wrapper can get you up and running with this scheme in no time. I'd be interested to hear folks' high-level thoughts on how D2D2T (or, more generally, data migration) configuration should work in Amanda. Should migration be a separate step from dumping (a la amflush)? Should migration be scheduled somehow (and if so, with what sort of parameters), or should Amanda require explicit instructions on which dumps to migrate and when? We're still laying the low-level groundwork for migration, so there's lots of time to pontificate. One usage of data migrations is archiving which is being used and often implemented as separate config doing only 0 dumps. I think that migration should be done in similar way i.e. separated from dumping - ammigrating with it's own configuration, with parameters selecting images/tapes from amanda configurations. - config,host,disk,level,frequency - tape support Of course the is only one aspect of migrations but important one that would add much value to amanda. Gunnar Gunnarsson
Archiving with amanda revisited
Hi all, Thanks for an excellent software. I was searching the forum and found a question regarding archiving. As I'm using vtapes this would be a great feature to have, being able to collect the 0 images and save them on tape and continue to use the tool kit, amrecover etc. What is the status on this useful feature ? http://archives.zmanda.com/amanda-archives/viewtopic.php?t=3866highligh t=archiving Gunnar Gunnarsson
Portrange setting 2.5.2p1
If portrange settings are changed in the server part with --with-tcpportrange=5,50100 --with-udpportrange=840,860 how does it effect the clients - do they need to be recompiled as well ? Thanks for your help Gunnar
