Results 1 to 3 of 3

Thread: amandad dumping core

  1. #1

    Default amandad dumping core

    Hi all,

    I just started an upgrade from a really old version of Amanda (2.4.3) to the latest stable version (2.5.1p2). The server still runs the old version, a few clients are running the new version and are having trouble with the "dump" backup type.

    On the server, I have this in the logs:

    FAIL dumper h20.foo.com /usr/local 20070214 0 [data timeout]
    sendbackup: start [h20.foo.com:/usr/local level 0]
    sendbackup: info BACKUP=/sbin/dump
    sendbackup: info RECOVER_CMD=/usr/bin/gzip -dc |/sbin/restore -f - ...
    sendbackup: info COMPRESS_SUFFIX=.gz
    sendbackup: info end
    | DUMP: Date of this level 0 dump: Wed Feb 14 00:57:02 2007
    | DUMP: Date of last level 0 dump: the epoch
    | DUMP: Dumping /dev/da0s1f (/usr/local) to standard output
    | DUMP: mapping (Pass I) [regular files]
    | DUMP: mapping (Pass II) [directories]
    | DUMP: estimated 138795 tape blocks.
    | DUMP: dumping (Pass III) [directories]
    | DUMP: dumping (Pass IV) [regular files]
    ? dumper: strange [missing size line from sendbackup]
    ? dumper: strange [missing end line from sendbackup]

    On the client, I have no sendbackup log for /usr/local, but this is the tail end of the last amandad debug log:

    amandad: debug 1 pid 73617 ruid 2 euid 2: start at Wed Feb 14 00:57:02 2007
    security_getdriver(name=BSD) returns 0x28093340
    amandad: version 2.5.1p2
    amandad: build: VERSION="Amanda-2.5.1p2"
    amandad: BUILT_DATE="Tue Jan 9 22:46:05 EST 2007"
    amandad: BUILT_MACH="FreeBSD screwdriver.foo.com 4.11-RELEASE-p9 Free
    BSD 4.11-RELEASE-p9 #10: Sun May 29 18:48:35 EDT 2005 [email]spork@devel2.foo.com[/email]:/
    usr/obj/usr/src/sys/PIII-XEON-single2 i386"
    amandad: CC="cc"
    amandad: CONFIGURE_COMMAND="'./configure' '--libexecdir=/usr/local/libexe
    c/amanda' '--with-amandahosts' '--with-fqdn' '--with-bsdtcp-security' '--with-du
    mp-honor-nodump' '--with-buffered-dump' '--disable-libtool' '--prefix=/usr/local
    ' '--with-user=operator' '--with-group=operator' '--with-index-server=screwdrive
    r' '--with-tape-server=screwdriver' '--with-config=DailySet1' '--with-udpportran
    ge=900,1000' '--with-portrange=30000,30100' '--with-gnutar-listdir=/usr/local/va
    r/amanda/gnutar-lists' '--with-gnutar=/usr/local/bin/gtar' '--without-server' '-
    -prefix=/usr/local' '--build=i386-portbld-freebsd4.11'"
    amandad: paths: bindir="/usr/local/bin" sbindir="/usr/local/sbin"
    amandad: libexecdir="/usr/local/libexec/amanda"
    amandad: mandir="/usr/local/man" AMANDA_TMPDIR="/tmp/amanda"
    amandad: AMANDA_DBGDIR="/tmp/amanda"
    amandad: CONFIG_DIR="/usr/local/etc/amanda" DEV_PREFIX="/dev/"
    amandad: RDEV_PREFIX="/dev/r" DUMP="/sbin/dump"
    amandad: RESTORE="/sbin/restore" VDUMP=UNDEF VRESTORE=UNDEF
    amandad: XFSDUMP=UNDEF XFSRESTORE=UNDEF VXDUMP=UNDEF VXRESTORE=UNDEF
    amandad: SAMBA_CLIENT=UNDEF GNUTAR="/usr/local/bin/gtar"
    amandad: COMPRESS_PATH="/usr/bin/gzip"
    amandad: UNCOMPRESS_PATH="/usr/bin/gzip" LPRCMD="/usr/bin/lpr"
    amandad: MAILER="/usr/bin/Mail"
    amandad: listed_incr_dir="/usr/local/var/amanda/gnutar-lists"
    amandad: defs: DEFAULT_SERVER="screwdriver" DEFAULT_CONFIG="DailySet1"
    amandad: DEFAULT_TAPE_SERVER="screwdriver" HAVE_MMAP HAVE_SYSVSHM
    amandad: LOCKING=POSIX_FCNTL DEBUG_CODE AMANDA_DEBUG_DAYS=4
    amandad: BSD_SECURITY RSH_SECURITY USE_AMANDAHOSTS
    amandad: CLIENT_LOGIN="operator" FORCE_USERID HAVE_GZIP
    amandad: COMPRESS_SUFFIX=".gz" COMPRESS_FAST_OPT="--fast"
    amandad: COMPRESS_BEST_OPT="--best" UNCOMPRESS_OPT="-dc"
    amandad: time 0.000: dgram_recv(dgram=0x28094324, timeout=0, fromaddr=0x280a4310
    )
    amandad: time 0.000: (sockaddr_in *)0x280a4310 = { 2, 904, x.x.4.28 }
    security_handleinit(handle=0x8051000, driver=0x28093340 (BSD))
    amandad: time 0.001: accept recv REQ pkt:
    <<<<<
    SERVICE sendbackup
    OPTIONS features=fffffeff9f00;hostname=h20.foo.com;
    DUMP /usr/local 0 1970:1:1:0:0:0 OPTIONS |;auth=bsd;compress-fast;index;
    >>>>>
    amandad: time 0.001: creating new service: sendbackup
    OPTIONS features=fffffeff9f00;hostname=h20.foo.com;
    DUMP /usr/local 0 1970:1:1:0:0:0 OPTIONS |;auth=bsd;compress-fast;index;

    amandad: time 0.002: sending ACK pkt:
    <<<<<
    >>>>>
    amandad: dgram_send_addr(addr=0xbfbff7f8, dgram=0x28094324)
    amandad: time 0.002: (sockaddr_in *)0xbfbff7f8 = { 2, 904, x.x.4.28 }
    amandad: dgram_send_addr: 0x28094324->socket = 0
    security_streaminit(stream=0x8070000, driver=0x28093340 (BSD))
    amandad: try_socksize: send buffer size is 65536
    amandad: try_socksize: receive buffer size is 65536
    amandad: time 0.008: stream_server: waiting for connection: 0.0.0.0.4855
    security_streaminit(stream=0x8079000, driver=0x28093340 (BSD))
    amandad: try_socksize: send buffer size is 65536
    amandad: try_socksize: receive buffer size is 65536
    amandad: time 0.008: stream_server: waiting for connection: 0.0.0.0.4216
    security_streaminit(stream=0x8082000, driver=0x28093340 (BSD))
    amandad: try_socksize: send buffer size is 65536
    amandad: try_socksize: receive buffer size is 65536
    amandad: time 0.008: stream_server: waiting for connection: 0.0.0.0.4276
    amandad: time 0.008: sending REP pkt:
    <<<<<
    CONNECT DATA 4855 MESG 4216 INDEX 4276
    OPTIONS features=fffffeff9ffeffffff7f;
    >>>>>
    amandad: dgram_send_addr(addr=0xbfbff878, dgram=0x28094324)
    amandad: time 0.009: (sockaddr_in *)0xbfbff878 = { 2, 904, x.x.4.28 }
    amandad: dgram_send_addr: 0x28094324->socket = 0
    amandad: time 0.086: dgram_recv(dgram=0x28094324, timeout=0, fromaddr=0x280a4310
    )
    amandad: time 0.086: (sockaddr_in *)0x280a4310 = { 2, 904, x.x.4.28 }
    amandad: time 0.086: received ACK pkt:
    <<<<<
    >>>>>
    amandad: time 0.252: stream_accept: connection from x.x.4.28.30012
    amandad: try_socksize: send buffer size is 65536
    amandad: try_socksize: receive buffer size is 65536
    amandad: time 0.425: stream_accept: connection from x.x.4.28.30013
    amandad: try_socksize: send buffer size is 65536
    amandad: try_socksize: receive buffer size is 65536
    amandad: time 0.609: stream_accept: connection from x.x.4.28.30014
    amandad: try_socksize: send buffer size is 65536
    amandad: try_socksize: receive buffer size is 65536
    security_close(handle=0x8051000, driver=0x28093340 (BSD))
    security_stream_seterr(0x8070000, write error on stream 4855: Broken pipe)
    amandad: time 791.368: sending NAK pkt:
    <<<<<
    ERROR write error on stream 4855: write error on stream 4855: Broken pipe
    >>>>>

    Additionally, someone else has opened a bug with the FreeBSD port maintainer, but the bug seems to be going nowhere:

    [url]http://www.freebsd.org/cgi/query-pr.cgi?pr=106685&cat=ports[/url]

    Any ideas on this one? I'd love to get "current" with amanda, but this is a pretty serious problem. I've seen this on the last dump for each host running the new client.

    If I could figure out how to let amandad drop a core file, I'd be happy to gdb it...

    Thanks,

    Charles

  2. #2

    Default Similar problem

    I am getting the same kind of error from amdump log.

    amandad: time 1943.783: dgram_send_addr(addr=0x807b258, dgram=0xb7f4a9e4)
    amandad: time 1943.783: (sockaddr_in *)0x807b258 = { 2, 936, 192.168.0.10 }
    amandad: time 1943.783: dgram_send_addr: 0xb7f4a9e4->socket = 0
    amandad: time 1943.783: security_close(handle=0x807b238, driver=0xb7f48a40 (BSD))
    amandad: time 1943.783: bsd: close handle '000-00000001'
    amandad: time 4245.110: security_stream_seterr(0x808b908, write error on stream 42165: Broken pipe)
    amandad: time 4245.110: sending NAK pkt:
    <<<<<
    ERROR write error on stream 42165: write error on stream 42165: Broken pipe
    >>>>>
    amandad: time 4245.110: closing service: /usr/lib/amanda/sendbackup
    amandad: time 4245.110: security_stream_close(0x808b908)
    amandad: time 4245.110: security_stream_close(0x8093940)
    amandad: time 4245.110: security_stream_close(0x809b978)
    amandad: time 4265.111: pid 18966 finish time Wed Aug 29 03:11:06 2007

    I am using 2.5.2p1 on both client and server. In fact the server is the client. It only fails for one of the disk list entries. The largest one. The size of the disk is about 140 G. Before this disk worked just fine. I had recently copied a lot of files on it so it grew in size. The disk has special file system. It is OCFS2 and used simultaneously by two computers. Thus it is a little slower than a built in disk.

    Another thing I did is that I set calcsize in dumptype to "server" because it was taking too long to estimate the disk. Now it does not take too long but it fails. The estimate is very inaccurate because it only uses data from last run and as I mentioned the disk grew in size but I am using very large tapes ( about 400 G each ) and I thought Amanda will just write data and correct the estimate on the next run.

    Anyways if someone can shed more light on this broken pipe problem I'll be very thankful. Or could someone explain to me what exactly does the "pipe" mean in this context. Is it the connection between the server and amandad or is it the pipe though which tar gets its input.

  3. #3
    Join Date
    Oct 2005
    Location
    Bay Area, CA
    Posts
    124

    Default

    Hi,

    is there any error in sendbackup.*.debug log?
    I would also try to increase dtimeout in amanda.conf.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •