head 1.2; access; symbols pkgsrc-2013Q2:1.2.0.20 pkgsrc-2013Q2-base:1.2 pkgsrc-2012Q4:1.2.0.18 pkgsrc-2012Q4-base:1.2 pkgsrc-2011Q4:1.2.0.16 pkgsrc-2011Q4-base:1.2 pkgsrc-2011Q2:1.2.0.14 pkgsrc-2011Q2-base:1.2 pkgsrc-2009Q4:1.2.0.12 pkgsrc-2009Q4-base:1.2 pkgsrc-2008Q4:1.2.0.10 pkgsrc-2008Q4-base:1.2 pkgsrc-2008Q3:1.2.0.8 pkgsrc-2008Q3-base:1.2 cube-native-xorg:1.2.0.6 cube-native-xorg-base:1.2 pkgsrc-2008Q2:1.2.0.4 pkgsrc-2008Q2-base:1.2 pkgsrc-2008Q1:1.2.0.2 pkgsrc-2008Q1-base:1.2 pkgsrc-2007Q4:1.1.0.8 pkgsrc-2007Q4-base:1.1 pkgsrc-2007Q3:1.1.0.6 pkgsrc-2007Q3-base:1.1 pkgsrc-2007Q2:1.1.0.4 pkgsrc-2007Q2-base:1.1 pkgsrc-2007Q1:1.1.0.2 pkgsrc-2007Q1-base:1.1; locks; strict; comment @# @; 1.2 date 2008.02.15.16.23.09; author gdt; state dead; branches; next 1.1; 1.1 date 2007.01.05.13.20.54; author obache; state Exp; branches; next ; desc @@ 1.2 log @Update to 2.5.2p1, provided by Nick Goffee of BBN. [bugfixes and minor improvements omitted; see NEWS] Changes in release 2.5.2 * krb5 auth is working * Works with IPv6 address * Amanda configuration file changes o amanda.conf changes + debugging is enabled in the config file see all debug_* config option + tapetype 'readblocksize', if maxtapeblocksize is set too larger for your hardware * Amanda command changes o amadmin: new 'holding list' and 'holding delete' subcommand. Changes in release 2.5.1p3 * Works with tar-1.16 and exit status of 1. Changes in release 2.5.1p2 * amoverview is working * dumptype starttime is working * Amanda command changes o amtape accept the -o arguments o amgetconf --list to list all tapetype, dumptype, holdingdisk or interface o amgetconf can return a value of a specific tapetype, dumptype, holdingdisk or interface Changes in release 2.5.1p1 * Remove contrib/sst. Changes in release 2.5.1 * Defects found by Coverity scan and Klocwork K7 analysis tools fixed. * Works with GNU tar 1.15.91 - work with new gtar state file format. * Open SSL encryption support * Two new authentication methods: bsdtcp, bsdudp. * Unlimited number of DLEs on a client with bsdtcp, rsh and ssh authentication methods. * Recovery process amrecover uses Secure API. amoldrecover command (same syntax and functionality as amrecover command) is provided for compatibility with old Amanda releases. amoldrecover command uses old amidxtaped/amindexd protocol. * Amanda debug files are separated into client/server/amandad and are also classified based on Amanda configuration name. * Amanda command changes o amfetchdump -o is replaced by -O. o amcheck -w option does all tests including the tape writable test. Use amcheck -t -w to do only the tape writable test. o -o command option to override Amanda configuration. See amanda man page for details. o amgetconf command doesn't write the BUGGY message when a entry is not found in the configuration file. * Amanda configuration file changes o amanda.conf changes + amrecover_do_fsf in amanda.conf defaults to yes + amrecover_check_label in amanda.conf defaults to yes + usetimestamps in amanda.conf to support multiple backup runs in a calendar day. + holdingdisk in amanda.conf supports new values: NEVER, AUTO, REQUIRED. + amandad_path, client_username and ssh_keys in amanda.conf for ssh/rsh authentication. o New amanda client configuration file - amanda-client.conf. Different client configuration file can be used for each Amanda configuration. + gnutar_list-dir and amandates can be specified in Amanda client configuration file - amanda-client.conf o .amandahosts format changes to allow use of secure API for recovery. o Amanda service entries in xinetd configuration has changed. Changes in release 2.5.0p1 * Add the 'amtape update' command. Changes in release 2.5.0 * Communication security/authentication: Kerberos 4/5, OpenSSH * Data security: Symmetric/Assymetric encrytion algorithms (aesutil and gpg encryption), Encryption can be done on server or client, Custom encryption utilities can be used. * Compression: Ability to add custom compression tools. This is a really useful feature especially since it can specified for a DLE. You can use different compression algorithm for images, binaries, ascii files and so on. * Dump images spanning multiple media volumes - Dump images are no longer restricted to a single media volume (tape or vtape). Data restoration can be done using amrecover and amfetchdump commands. * Auto tape labelling - This optional feature is good for disk backups. Change in release 2.4.5 * new displayunit global option to select the unit use to display number k=kilo, m=mega, g=giga, t=tera. * new amoverview -skipmissed option. Change in release 2.4.5b1 * holding disk disk use timestamped directory. * autoflush flush today's dump. * new bumppercent global option, this should improve bumping criteria, the bumpsize is set to a fixed value which can be adequate for small and large disk. * bumpsize, bumppercent, bumpdays and bumpmult can be in a dumptype. * calcsize support include and exclude like gnutar. * new 'estimate' dumptype option to select estimate type: CLIENT: estimate by the dumping program. CALCSIZE: estimate by the calcsize program, a lot faster but less acurate. SERVER: estimate based on statistic from previous run, take second but can be wrong on the estimate size. @ text @$NetBSD: patch-an,v 1.1 2007/01/05 13:20:54 obache Exp $ Prevent amverify from looping forever. Patch provided by Jean-Louis Martineau on amanda-users: http://groups.yahoo.com/group/amanda-users/message/55099 --- restore-src/amrestore.c.orig 2003-02-10 15:22:11.000000000 +0000 +++ restore-src/amrestore.c @@@@ -482,6 +482,7 @@@@ char **argv; char *e; char *err; char *label = NULL; + int count_error; for(fd = 3; fd < FD_SETSIZE; fd++) { /* @@@@ -667,25 +668,29 @@@@ char **argv; get_pname()); } - while(file.type == F_TAPESTART || file.type == F_DUMPFILE) { - amfree(filename); - filename = make_filename(&file); + count_error=0; + while(count_error < 10) { + if(file.type == F_TAPEEND) break; found_match = 0; - for(me = match_list; me; me = me->next) { - if(disk_match(&file,me->datestamp,me->hostname,me->diskname) != 0) { - found_match = 1; - break; + if(file.type == F_DUMPFILE) { + amfree(filename); + filename = make_filename(&file); + for(me = match_list; me; me = me->next) { + if(disk_match(&file,me->datestamp,me->hostname,me->diskname) != 0) { + found_match = 1; + break; + } + } + fprintf(stderr, "%s: %3d: %s ", + get_pname(), + file_number, + found_match ? "restoring" : "skipping"); + if(file.type != F_DUMPFILE) { + print_header(stderr, &file); + } else { + fprintf(stderr, "%s\n", filename); } } - fprintf(stderr, "%s: %3d: %s ", - get_pname(), - file_number, - found_match ? "restoring" : "skipping"); - if(file.type != F_DUMPFILE) { - print_header(stderr, &file); - } else { - fprintf(stderr, "%s\n", filename); - } if(found_match) { restore(&file, filename, isafile); if(compress_pid > 0) { @@@@ -718,6 +723,7 @@@@ char **argv; if((tapedev = tape_open(tapename, 0)) < 0) { error("could not open %s: %s", tapename, strerror(errno)); } + count_error++; } else { /* * If the last read got something (even an error), we can @@@@ -726,6 +732,7 @@@@ char **argv; if(tapefd_fsf(tapedev, 1) < 0) { error("could not fsf %s: %s", tapename, strerror(errno)); } + count_error=0; } file_number++; read_file_header(&file, isafile); @ 1.1 log @Prevent amverify from looping forever. Patch provided by Jean-Louis Martineau on amanda-users: http://groups.yahoo.com/group/amanda-users/message/55099 Pointed out by Jukka Salmi in PR 35361. @ text @d1 1 a1 1 $NetBSD$ @