Bug#1059442: gitg: Commit dialog overrides Ctrl+Left/Right
Package: gitg Version: 41-2 Severity: important Tags: patch upstream X-Debbugs-Cc: stefan.tau...@gmx.at Dear Maintainers, after upgrading to Bookworm I have noticed that Ctrl+Left/Right does not work as expected in the commit dialog of gitg. Usually Ctrl+Left/Right moves the cursor one word to the left or right, respectively. This is true for GUI text boxes like the one in reportbug I am currently typing in but also terminal applications etc. The version of gitg in Bookworm has added a commit message history feature to the commit dialog that is navigated via said shortcuts, unfortunately, and it drives me a little bit crazy. This problem has been reported (1) and fixed (2) upstream by using Alt+Page up/Page down. I have added the upstream patch to the bookworm branch of my salsa fork of gitg (3). "debuild -b -uc -us" isn't working due to an unrelated lintian error: "E: gitg: library-not-linked-against-libc [usr/lib/gitg/gitg/plugins/libdiff.so]" But "fakeroot debian/rules binary" spat out a working .deb that seems to work fine. I believe this is a candidate for being backported to Bookworm as the source change is tiny and the bug is a regression introduced by Bookworm. I am happy to help if need be (but I am not a Debian maintainer). KR 1: https://gitlab.gnome.org/GNOME/gitg/-/issues/351 2: https://gitlab.gnome.org/GNOME/gitg/-/commit/d768b107aef1944d945e73ecf14d74746338afe4 3: https://salsa.debian.org/stefanct/gitg/-/commit/bdb80f0a76c789cdb3a78d5e4270f7d3daa4f67f -- System Information: Debian Release: 12.4 APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'stable-security'), (500, 'stable'), (89, 'testing'), (10, 'unstable') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 6.1.0-13-amd64 (SMP w/4 CPU threads; PREEMPT) Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages gitg depends on: ii dbus-user-session [default-dbus-session-bus] 1.14.10-1~deb12u1 ii dbus-x11 [dbus-session-bus] 1.14.10-1~deb12u1 ii dconf-gsettings-backend [gsettings-backend] 0.40.0-4 ii gir1.2-peas-1.0 1.34.0-1+b1 ii git 1:2.39.2-1.1 ii gsettings-desktop-schemas 43.0-1 ii libc6 2.36-9+deb12u3 ii libcairo2 1.16.0-7 ii libdazzle-1.0-0 3.44.0-1 ii libgdk-pixbuf-2.0-0 2.42.10+dfsg-1+b1 ii libgee-0.8-2 0.20.6-1 ii libgirepository-1.0-1 1.74.0-3 ii libgit2-glib-1.0-01.1.0-1+b1 ii libglib2.0-0 2.74.6-2 ii libgspell-1-2 1.12.0-1+b2 ii libgtk-3-03.24.38-2~deb12u1 ii libgtksourceview-4-0 4.8.4-4 ii libjson-glib-1.0-01.6.6-1 ii libpango-1.0-01.50.12+ds-1 ii libpeas-1.0-0 1.34.0-1+b1 ii libsecret-1-0 0.20.5-3 ii libxml2 2.9.14+dfsg-1.3~deb12u1 gitg recommends no packages. gitg suggests no packages. -- no debconf information
Bug#1034655: lightdm: Login dialog after resuming from standby should match xflock4
Package: lightdm Version: 1.26.0-8 Severity: normal X-Debbugs-Cc: stefan.tau...@gmx.at Dear Maintainer, since I am using autologin (because I am using FDE and have to unlock the OS before boot anyway) there are usually only two ways to interact with the lightdm: after locking the session with xflock4 and after resuming from standby. While the UI behaves as I which after locking manually it does not when resuming: I'd like to simply enter my password to unlock the session after resume but the UI is blank in that case instead of having the username predefined and the focus on the password field. Objectively, this is a minor thing but for me it's about 50% of my interactions :) It's also a regression IMhO - at least in Debian 10 this worked perfectly (I kinda skipped 11). I presume the behavior is due to lightdm itself but maybe it is because of how the session gets locked when entering suspend (which is via systemctl suspend)? The only changes I made to lightdm.conf are setting autologin-user to my username and autologin-user-timeout=0. This is a pretty fresh install of Bookworm. KR -- System Information: Debian Release: 12.0 APT prefers testing-security APT policy: (500, 'testing-security'), (500, 'testing') Architecture: amd64 (x86_64) Kernel: Linux 6.1.0-7-amd64 (SMP w/16 CPU threads; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages lightdm depends on: ii adduser3.132 ii dbus 1.14.6-1 ii debconf [debconf-2.0] 1.5.82 ii libaudit1 1:3.0.9-1 ii libc6 2.36-9 ii libgcrypt201.10.1-3 ii libglib2.0-0 2.74.6-2 ii libpam-systemd [logind]252.6-1 ii libpam0g 1.5.2-6 ii libxcb11.15-1 ii libxdmcp6 1:1.1.2-3 ii lightdm-gtk-greeter [lightdm-greeter] 2.0.8-2+b1 ii sysvinit-utils [lsb-base] 3.06-4 Versions of packages lightdm recommends: ii xserver-xorg 1:7.7+23 Versions of packages lightdm suggests: pn accountsservice ii upower 0.99.20-2 pn xserver-xephyr -- Configuration Files: /etc/lightdm/lightdm.conf changed [not included] -- debconf information: lightdm/daemon_name: /usr/sbin/lightdm * shared/default-x-display-manager: lightdm
Bug#967933: marked as pending in ant
On Mon, 11 Jul 2022 00:06:35 + Tony Mancill wrote: > Control: tag -1 pending > > Hello, > > Bug #967933 in ant reported by you has been fixed in the > Git repository and is awaiting an upload. You can see the commit > message below and you can check the diff of the fix at: > > https://salsa.debian.org/java-team/ant/-/commit/b15414ef50d903860f6264e0b49d4658231aa3d5 Hi Tony, thanks! What do you think about the junit5 dependencies mentioned earlier that I have included in my attempt (https://salsa.debian.org/stefanct/ant/-/commit/3ca052c0b7a1c7300659a791ab18989af6fc25bd)? I forgot about everything I did back then TBH though. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#967933: ant-optional: Missing ant-junitlauncher.jar
Ok, got it. I missed some additional build requirements for Ant on Junit5 if one wants to build junitlauncher completely and the whole confined stuff threw me off. Interestingly, junit5 is then not needed at runtime anymore - at least not for how josm uses it. I have not tested other applications though. I have made some additional useful changes but I do not intend to do a NMU (yet) - it would be my first and I am not really confident that this does not break anything - unless instructed to do so. Feel free to copy any of it: https://salsa.debian.org/stefanct/ant/-/commits/master -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#967933: ant-optional: Missing ant-junitlauncher.jar
On Wed, 05 Aug 2020 09:55:57 +0200 Matthijs Kooijman wrote: > > If my analysis is correct, could you make the changes so > ant-junitlauncher is included? Hello, the good news is, adding that module is not a big deal. I think I have done that correctly. Even better, it gets rid of the specific error you reported! However, that does not make josm test successfully for me. I get another message that could either mean I built ant-optional not correctly or there is something weird going on in josm: test: [echo] Running unit tests with JUnit [junitlauncher] Error: Could not find or load main class org.apache.tools.ant.taskdefs.optional.junitlauncher.StandaloneLauncher [junitlauncher] Caused by: java.lang.ClassNotFoundException: org.apache.tools.ant.taskdefs.optional.junitlauncher.StandaloneLauncher The documentation of StandaloneLauncher hints at it should not actually be used externally. However, it is referenced in JUnitLauncherTaskTest.java (which is part of the exposed/"confined" classes) in a test... it's a bit weird. Since there is some discussion about a newer ant release in some of the josm tickets with similar problems I tried updating to the currently available upstream releases (1.10.10 and 1.10.11) but it doesn't fix that issue. One of the devs pointed me to the CI configuration that builds on Ubuntu. Maybe there is some hint in there but I dont see it... https://github.com/openstreetmap/josm/blob/master/.github/workflows/ant.yml I am out of ideas. At least it builds fine and is executable... -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#969728: claws-mail-vcalendar-plugin: Incorrect timezone handling
Package: claws-mail-vcalendar-plugin Version: 3.17.3-2 Severity: important Dear Maintainer, with claws-mail from Buster (3.17.3-2) the vcal plugin seems to miscalculate some timestamps due to insufficient timezone handling. This has been addressed a few times in the past but I think the following .ics triggers something differently since all reports (of fixed bugs) that I could find are way older than my version. BEGIN:VCALENDAR VERSION:2.0 BEGIN:VEVENT CREATED:20200901T141753Z DESCRIPTION:blabla\nmore bla\n DTEND;TZID=Europe/Vienna:20200907T14 DTSTART;TZID=Europe/Vienna:20200907T13 LOCATION:https://some.url/bla?blub SUMMARY;LANGUAGE=us-EN:some text UID:{dfdf450b-8eea-457b-9a73-8aa1d5bf1e9b} BEGIN:VALARM TRIGGER:-PT10M ACTION:DISPLAY DESCRIPTION:Reminder END:VALARM END:VEVENT END:VCALENDAR It is displayed at - starting at "Mon, 7 Sep 2020 15:00:00 CEST" - ending at "Mon, 7 Sep 2020 16:00:00 CEST" Since my local time is according to the Europe/Vienna timezone just like in the .ics (which is correctly displayed as CEST) the correct time would be 13:00 to 14:00. It seems like the DST is applied twice or something? -- System Information: Debian Release: 10.5 APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'oldstable-updates'), (500, 'stable'), (91, 'testing'), (10, 'unstable'), (5, 'oldstable') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 5.5.0-0.bpo.2-amd64 (SMP w/4 CPU cores) Kernel taint flags: TAINT_WARN, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages claws-mail-vcalendar-plugin depends on: ii claws-mail 3.17.3-2 ii libatk1.0-0 2.30.0-2 ii libc62.28-10 ii libcairo21.16.0-4 ii libcurl3-gnutls 7.64.0-4+deb10u1 ii libdb5.3 5.3.28+dfsg1-0.5 ii libetpan20 1.9.3-2 ii libexpat12.2.6-2+deb10u1 ii libfontconfig1 2.13.1-2 ii libfreetype6 2.9.1-3+deb10u1 ii libgdk-pixbuf2.0-0 2.38.1+dfsg-1 ii libglib2.0-0 2.58.3-2+deb10u2 ii libgnutls30 3.6.7-4+deb10u5 ii libgtk2.0-0 2.24.32-3 ii libical3 3.0.4-3 ii liblockfile1 1.14-1.1 ii libpango-1.0-0 1.42.4-8~deb10u1 ii libpangocairo-1.0-0 1.42.4-8~deb10u1 ii libpangoft2-1.0-01.42.4-8~deb10u1 ii libsasl2-2 2.1.27+dfsg-1+deb10u1 claws-mail-vcalendar-plugin recommends no packages. claws-mail-vcalendar-plugin suggests no packages. -- no debconf information
Bug#958929: git: Regression due to CVE-2020-11008 fix
Package: git Version: 1:2.20.1-2+deb10u3 Severity: normal Dear Maintainer, the vulnerability in CVE-2020-11008 is related to the handling of credential helpers in git. In Buster this has been fixed in 1:2.20.1-2+deb10u3. This broke my existing configuration where repositories have credential.helper=store set. This is documented in /usr/share/man/man1/git-credential-store.1.gz and other files from git, git-doc etc. I am unsure how to proceed... is this helper now unsupported? Is this a simple regression that should be fixed? Do other alternatives like git-credential-cache still work or are they broken as well? -- System Information: Debian Release: 10.3 APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'oldstable-updates'), (500, 'stable'), (91, 'testing'), (10, 'unstable'), (5, 'oldstable') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 5.3.0-0.bpo.2-amd64 (SMP w/4 CPU cores) Kernel taint flags: TAINT_WARN, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages git depends on: ii git-man 1:2.20.1-2+deb10u3 ii libc62.28-10 ii libcurl3-gnutls 7.64.0-4+deb10u1 ii liberror-perl0.17027-2 ii libexpat12.2.6-2+deb10u1 ii libpcre2-8-0 10.32-5 ii perl 5.28.1-6 ii zlib1g 1:1.2.11.dfsg-1 Versions of packages git recommends: ii ca-certificates 20190110 ii less 487-0.1+b1 ii openssh-client [ssh-client] 1:7.9p1-10+deb10u2 ii patch2.7.6-3+deb10u1 Versions of packages git suggests: ii gettext-base 0.19.8.1-9 ii git-cvs 1:2.20.1-2+deb10u3 pn git-daemon-run | git-daemon-sysvinit ii git-doc 1:2.20.1-2+deb10u3 pn git-el ii git-email 1:2.20.1-2+deb10u3 pn git-gui pn git-mediawiki ii git-svn 1:2.20.1-2+deb10u3 ii gitk 1:2.20.1-2+deb10u3 pn gitweb -- no debconf information
Bug#940992: firefoxdriver: does not work with firefox > 47 at all
On Mon, 23 Sep 2019 12:56:48 +0200 Sascha Girrulat wrote: > Hi, > > the error below does not depend to firefox driver. The firefoxdriver ist > a leftover for older firefox versions. The package documentation[3] > describes how to enable the geckodriver and there is a bug[1] (#874207) > filed for firefox too. > > The packge firefoxdriver will be removed with the next update of > python-selenium in a few days[2]. > > Regards > Sascha > > [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=874207 > [2] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=939398 > [3] /usr/share/doc/python-selenium/README.Debian Hi, [3] is not installed by python3-selenium (nor an equivalent). The respective NEWS.Debian.gz of python3-selenium refers to that non-existent README.Debian though. It basically suggests fetching the binary from https://github.com/mozilla/geckodriver/releases and install it in a directory in $PATH. I can confirm that this works on Debian Buster (python3-selenium 3.14.1+dfsg1-1) with the latest geckodriver (v0.26.0 from 2019-10-12). What's the purpose of this bug? If it shouldn't be closed, shouldn't it depend on #874207? -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#573254: (no subject)
On Thu, 24 Oct 2019 00:45:50 -0500 Michael Lustfield wrote: > This was reported against 1.3.1, however 1.4.2 is available in oldstable. Can > you confirm if this issue is still present? If so, would you also be able to > also check against the latest git revision (many bug fixes are not released)? > Hi, I have checked with the one in Debian Buster (1.4.2-1) and Bullseye (1.4.3-1). The same error is emitted for the du command. The code in git as of now does not look like it has been changed either: https://github.com/rsnapshot/rsnapshot/blob/3b3fc2c1676b5f93f33671dd07d6569ec95c4dd4/rsnapshot-program.pl#L1086 KR -- Dipl.-Ing. Stefan Tauner PhD Student @ https://ti.tuwien.ac.at/ecs/
Bug#947819: regression: iwlwifi hangs kernel completely sometimes
Package: src:linux Version: 4.19.67-2+deb10u2 Severity: important Dear Maintainer, my Intel Centrino 6205 in my Thinkpad T430 sometimes hangs the hole system with Linux 4.19. I *think* it always related to the hardware kill switch and reloading the driver afterwards, but I am not completely sure since my "analysis" was done on several lock up occasions over the period of a few months. I cannot reproduce it effectively but it has incurred data loss thus I have set the elevated importance in this bug report. The reason why I tinker with the kill switch and/or the module at all is that auto-reconnecting (with nm-manager) does not work consistently. Both, the non-working reconnect and the lock ups, are a regression to Debian 9 stretch where everything worked flawlessly. I have attached some kernel log snippets that I was able to gather that show some warnings related to iwlwifi but I am not 100% sure they are actually related to the lock ups. Since it's very hard to reproduce and it is quite annoying I will now switch to Buster's backport kernel (5.3) and see if that helps. KR -- Package-specific info: ** Version: Linux version 4.19.0-6-amd64 (debian-ker...@lists.debian.org) (gcc version 8.3.0 (Debian 8.3.0-6)) #1 SMP Debian 4.19.67-2+deb10u2 (2019-11-11) ** Command line: BOOT_IMAGE=/boot/vmlinuz-4.19.0-6-amd64 root=/dev/mapper/vg-root ro quiet ** Tainted: OE (12288) * Out-of-tree module has been loaded. * Unsigned module has been loaded. ** Model information sys_vendor: LENOVO product_name: 2349PT4 product_version: ThinkPad T430 chassis_vendor: LENOVO chassis_version: Not Available bios_vendor: LENOVO bios_version: G1ETB7WW (2.77 ) board_vendor: LENOVO board_name: 2349PT4 board_version: No DPK ** Loaded modules: iwldvm mac80211 iwlwifi cfg80211 btrfs zstd_compress zstd_decompress xxhash ufs qnx4 hfsplus hfs minix vfat msdos fat jfs xfs ctr ccm rfcomm fuse pci_stub vboxpci(OE) vboxnetadp(OE) vboxnetflt(OE) cmac cpufreq_userspace vboxdrv(OE) cpufreq_conservative cpufreq_powersave bnep binfmt_misc btusb btrtl btbcm btintel uvcvideo bluetooth videobuf2_vmalloc videobuf2_memops videobuf2_v4l2 videobuf2_common videodev drbg ansi_cprng ecdh_generic media intel_rapl msr arc4 snd_hda_codec_hdmi x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel mei_wdt snd_hda_codec_realtek snd_hda_codec_generic kvm snd_hda_intel irqbypass intel_cstate snd_hda_codec intel_uncore snd_hda_core snd_hwdep intel_rapl_perf pcspkr serio_raw wmi_bmof sg snd_pcm iTCO_wdt iTCO_vendor_support thinkpad_acpi snd_timer nvram tpm_tis tpm_tis_core snd mei_me tpm mei pcc_cpufreq soundcore rfkill ac battery rng_core evdev parport_pc ppdev lp parport sunrpc ip_tables x_tables autofs4 ext4 crc16 mbcache jbd2 fscrypto ecb algif_skcipher af_alg dm_crypt dm_mod raid10 raid1 raid0 multipath linear raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c crc32c_generic md_mod sd_mod crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel pcbc i915 ahci libahci aesni_intel libata aes_x86_64 crypto_simd cryptd glue_helper psmouse scsi_mod i2c_i801 sdhci_pci xhci_pci i2c_algo_bit cqhci xhci_hcd drm_kms_helper sdhci lpc_ich mmc_core ehci_pci ehci_hcd drm e1000e usbcore thermal usb_common wmi video button ** PCI devices: 00:00.0 Host bridge [0600]: Intel Corporation 3rd Gen Core processor DRAM Controller [8086:0154] (rev 09) Subsystem: Lenovo 3rd Gen Core processor DRAM Controller [17aa:21f3] Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx- Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- SERR- Kernel driver in use: ivb_uncore 00:02.0 VGA compatible controller [0300]: Intel Corporation 3rd Gen Core processor Graphics Controller [8086:0166] (rev 09) (prog-if 00 [VGA controller]) Subsystem: Lenovo 3rd Gen Core processor Graphics Controller [17aa:21f3] Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+ Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- SERR- Kernel driver in use: i915 Kernel modules: i915 00:14.0 USB controller [0c03]: Intel Corporation 7 Series/C210 Series Chipset Family USB xHCI Host Controller [8086:1e31] (rev 04) (prog-if 30 [XHCI]) Subsystem: Lenovo 7 Series/C210 Series Chipset Family USB xHCI Host Controller (ThinkPad T430) [17aa:21f3] Control: I/O- Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+ Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- SERR- Kernel driver in use: xhci_hcd Kernel modules: xhci_pci 00:16.0 Communication controller [0780]: Intel Corporation 7 Series/C216 Chipset Family MEI Controller #1 [8086:1e3a] (rev 04) Subsystem: Lenovo 7 Series/C216 Chipset Family MEI Controller [17aa:21f3] Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Bug#945782: python(3)-libtmux should depend on tmux package
Package: python-libtmux Version: 0.8.0-1 Severity: normal Hi, when tmux is not installed one will encounter libtmux.exc.TmuxCommandNotFound exceptions as soon as one tries to do anything useful with the package. For example the following Python snipped provokes this behavior: import libtmux server = libtmux.Server() session = server.new_session(session_name="test") The same behavior can be observed with the Python 3 version. My conclusion is that both versions of the package should depend on tmux. KR -- System Information: Debian Release: 10.2 APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'oldstable-updates'), (500, 'stable'), (91, 'testing'), (10, 'unstable'), (5, 'oldstable') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 4.19.0-6-amd64 (SMP w/4 CPU cores) Kernel taint flags: TAINT_WARN, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages python-libtmux depends on: ii python 2.7.16-1 python-libtmux recommends no packages. python-libtmux suggests no packages. -- no debconf information
Bug#909041: xorg: Xorg log filling up with (EE) modeset(0): Failed to get GBM bo for flip to new front.
Hi, this is the upstream bug report: https://gitlab.freedesktop.org/xorg/xserver/issues/68 There is some discussion about the implementation to be found here: https://gitlab.freedesktop.org/xorg/xserver/merge_requests/131 The resulting patch has been merged: https://gitlab.freedesktop.org/xorg/xserver/merge_requests/193 It would be nice to get this backported to stable IMHO since running out of space on / (as just happened to me) is rather bad... ;) -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#942100: openssh-server: /etc/ssh/sshd_config unconditionally overwritten by update
Package: openssh-server Version: 1:7.9p1-10+deb10u1 Severity: important Hi, this just bit me on current stable (Buster) while updating from the security repo: The following packages will be upgraded: openssh-client (1:7.9p1-10 => 1:7.9p1-10+deb10u1) openssh-server (1:7.9p1-10 => 1:7.9p1-10+deb10u1) openssh-sftp-server (1:7.9p1-10 => 1:7.9p1-10+deb10u1) 3 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. Need to get 1.178 kB of archives. After this operation, 0 B of additional disk space will be used. Do you want to continue? [Y/n] Get:1 http://security.debian.org/debian-security buster/updates/main amd64 openssh-sftp-server amd64 1:7.9p1-10+deb10u1 [44,6 kB] Get:2 http://security.debian.org/debian-security buster/updates/main amd64 openssh-server amd64 1:7.9p1-10+deb10u1 [352 kB] Get:3 http://security.debian.org/debian-security buster/updates/main amd64 openssh-client amd64 1:7.9p1-10+deb10u1 [782 kB] Fetched 1.178 kB in 0s (4.945 kB/s) Reading changelogs... Done Preconfiguring packages ... (Reading database ... 498927 files and directories currently installed.) Preparing to unpack .../openssh-sftp-server_1%3a7.9p1-10+deb10u1_amd64.deb ... Unpacking openssh-sftp-server (1:7.9p1-10+deb10u1) over (1:7.9p1-10) ... Preparing to unpack .../openssh-server_1%3a7.9p1-10+deb10u1_amd64.deb ... Unpacking openssh-server (1:7.9p1-10+deb10u1) over (1:7.9p1-10) ... Preparing to unpack .../openssh-client_1%3a7.9p1-10+deb10u1_amd64.deb ... Unpacking openssh-client (1:7.9p1-10+deb10u1) over (1:7.9p1-10) ... Setting up openssh-client (1:7.9p1-10+deb10u1) ... Setting up openssh-sftp-server (1:7.9p1-10+deb10u1) ... Setting up openssh-server (1:7.9p1-10+deb10u1) ... Replacing config file /etc/ssh/sshd_config with new version rescue-ssh.target is a disabled or a static unit, not starting it. Processing triggers for man-db (2.8.5-2) ... Processing triggers for systemd (241-7~deb10u1) ... The important line is the forth from the bottom. Since I have changed the port of SSHD this makes it impossible to open new connections afterwards. I can't believe that making computers secure by essentially disconnecting their admins is the desired behavior of this package (update). Arguably, changing the port back to its default (as in my case) might even increase security risks. ;) AFAIK there is no way to override the settings from the standard config file (by files in a *.d directory as requested in other bug reports). If there is no other (well-documented) workaround I strongly consider this behavior a bug. -- System Information: Debian Release: 10.1 APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'stable'), (91, 'testing'), (10, 'unstable') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 4.19.0-6-amd64 (SMP w/4 CPU cores) Kernel taint flags: TAINT_WARN, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages openssh-server depends on: ii adduser3.118 ii debconf [debconf-2.0] 1.5.71 ii dpkg 1.19.7 ii libaudit1 1:2.8.4-3 ii libc6 2.28-10 ii libcom-err21.44.5-1+deb10u2 ii libgssapi-krb5-2 1.17-3 ii libkrb5-3 1.17-3 ii libpam-modules 1.3.1-5 ii libpam-runtime 1.3.1-5 ii libpam0g 1.3.1-5 ii libselinux12.8-1+b1 ii libssl1.1 1.1.1d-0+deb10u1 ii libsystemd0241-7~deb10u1 ii libwrap0 7.6.q-28 ii lsb-base 10.2019051400 ii openssh-client 1:7.9p1-10+deb10u1 ii openssh-sftp-server1:7.9p1-10+deb10u1 ii procps 2:3.3.15-2 ii ucf3.0038+nmu1 ii zlib1g 1:1.2.11.dfsg-1 Versions of packages openssh-server recommends: ii libpam-systemd [logind] 241-7~deb10u1 ii ncurses-term 6.1+20181013-2+deb10u1 ii xauth1:1.0.10-1 Versions of packages openssh-server suggests: pn molly-guard pn monkeysphere pn rssh pn ssh-askpass pn ufw -- debconf information excluded
Bug#725859: pdfchain: running the program fails with segmentation fault
Hi, this is still a problem in Debian Buster (current stable(!), still version 1:0.4.4.2-1). Launching it from the command line is enough to trigger the segfault. I did not try the patch yet or compiling from source and debugging. However, I found a workable (albeit slot) workaround: running pdfchain in valgrind ;) This bug has been open for a long time despite its severity. Is there still an active maintainer or should we file an "orphaned" bug because of that? -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#941911: g++: Broken -x/suffix handling leads to spurious -Wc++-compat warnings
Package: g++-8 Version: 8.3.0-6 Severity: normal Dear Maintainer, I noticed a bug in handling -Wc++-compat when (not) using the -x option. Suppose you have two files containing valid C saved in files a.c and b.c, for example, "int main(void) {}" in a.c and just a newline in b.c. According to the documentation AFAICT you should be able to compile this without errors in a single step by executing g++ -Wc++-compat a.c b.c since the correct language of the file should be determined automatically by the suffix. However, this leads to two times cc1plus: warning: command line option ‘-Wc++-compat’ is valid for C/ObjC but not for C++ One should be able to remove the warnings by forcing the language used to C by prepending "-x c" as in g++ -Wc++-compat -x c a.c b.c since the manual states that -x "applies to all following input files until the next -x option". However, this is not the case because only the first spurious warning is fixed by this and ony adding another -x c just before the second file, i.e., g++ -Wc++-compat -x c a.c -x c b.c works as expected without any warnings. This problem is not depending on any arch specifics (tested with native g++ 8.3.0 on amd64, and cross compilers for riscv64-linux-gnu, and arm-linux-gnueabihf). A quick test with g++ 6.3.0-18+deb9u1 on amd64 revealed the same behavior. I did not (and do not intend to but could if need be) report this upstream. KR, Stefan -- System Information: Debian Release: 10.1 APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'stable'), (91, 'testing'), (10, 'unstable') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 4.19.0-6-amd64 (SMP w/4 CPU cores) Kernel taint flags: TAINT_WARN, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages g++-8 depends on: ii gcc-88.3.0-6 ii gcc-8-base 8.3.0-6 ii libc62.28-10 ii libgmp10 2:6.1.2+dfsg-4 ii libisl19 0.20-2 ii libmpc3 1.1.0-1 ii libmpfr6 4.0.2-1 ii libstdc++-8-dev 8.3.0-6 ii zlib1g 1:1.2.11.dfsg-1 g++-8 recommends no packages. Versions of packages g++-8 suggests: pn g++-8-multilib ii gcc-8-doc 8.3.0-1~bpo10+1 pn libstdc++6-8-dbg -- no debconf information
Bug#918754: bash: $PATH in bash does not include /sbin and /usr/sbin
On Wed, 11 Sep 2019 14:18:32 + "Jakubith, Boris" wrote: > I think this no _not_ a good idea. The semantics of 'su' is correct. The > only error is that many users up to day count on the wrong behaviour. > […] > > You can set 'ALWAYS_SET_PATH yes' for your installation, but generally - in > a default install - this would be sooo wrong, especially because there many > alternatives. I can't argue with that, however I want to point out that the root terminal (at least on Mate) as described in the Debian wiki[1] is executed via "gksu /usr/bin/x-terminal-emulator" that lacks the correct (PATH) environment too (I did an upgrade so maybe this is a relic). Changing the command to use gksudo instead sets up the right directories AFAICT. This definitely looks like a bug to me. Can anybody confirm the behavior (of launching a root terminal in a desktop environment and not having the sbin directories in PATH) on a fresh Gnome and/or Mate install? If this is deemed a bug shall we repurpose this one or create a new one? Sidenote: It is also no longer possible to launch X applications from such a (root) terminal which used to work (can't remember if I had to persuade it to do so with some configuration files though). 1: https://wiki.debian.org/Root -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#896063: gcc-default-ports: cross compiler for RISC-V 32bit
On Fri, 20 Apr 2018 10:22:05 +0200 Aurelien Jarno wrote: > On 2018-04-20 12:40, Paul Wise wrote: > > On Fri, Apr 20, 2018 at 12:05 PM, Karsten Merker wrote: > > > > > I'm CCing the debian-riscv list in case somebody there should > > > have further insights regarding riscv32 support in glibc. > > > > I think Heinrich was looking for a bare-metal compiler for > > microcontroller-class RISC-V rather than one for a hypothetical Debian > > riscv32 port? > > In that case gcc-default-ports is probably the wrong package. Someone > has to package it as a different package similar to gcc-arm-none-eabi. So this should actually be an RFP class bug? Should this one be converted or closed and another one opened? -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#898516: cryptroot: verbosity of keyfile copy operation
Package: cryptsetup Version: 2:2.0.2-1 Hi, it took me quite some time to figure out how to set up my initrd to include the correct key file. The documentation is actually quite clear *when* one finds the correct bits (search engines find way too many outdated tutorials instead unfortunately... due to https://anonscm.debian.org/robots.txt). One thing that could have saved me some time (and looking into the actual source code of the cryptroot hook) would be a clear indication when key files are copied like it is done for other operations in initramfs scripts. I have expected something like "Adding keyfile ${KEYFILE} as /cryptroot-keyfiles/${node}.key" or similar when updating the initramfs using -v. I would have sent a patch, however, I am not entirely sure how to add a message to add_device() properly since it uses stdout/echo to return required kernel modules. I guess the best would be to refactor the function and use the generic copy_file() function of the hook-functions library that prints a suitable message? Also, since the name and destination directory of the key files are chosen by the script and not influenced by the source name/user it might be nice to document the (naming) scheme. The problem exists since the code was added in r1044 (cf. bug #786578): https://anonscm.debian.org/viewvc/pkg-cryptsetup/cryptsetup/trunk/debian/initramfs/cryptroot-hook?r1=1043=1044=1044; -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#850892: gcc should not warn about the assignment-allocation character 'm' when POSIX is enabled
On Wed, 11 Jan 2017 00:14:03 +0100 Matthias Klose <d...@debian.org> wrote: > Please could you forward that upstream yourself? Sure, upstream bug report is at https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79055 KR -- Dipl.-Ing. Stefan Tauner Research and Development Embedded Systems Department University of Applied Sciences Technikum Wien Hoechstaedtplatz 6, 1200 Vienna, Austria T: +43 1 333 40 77-316 E: stefan.tau...@technikum-wien.at I: embsys.technikum-wien.at I: www.technikum-wien.at
Bug#850892: gcc should not warn about the assignment-allocation character 'm' when POSIX is enabled
package: gcc version: 4:6.2.1-1 Hi, there was a previous bug report about the issue in question (#584511) but it was closed and archived because it was targeting gcc-4.4 which was removed from Debian (I tried to reopen it to no avail). However, the problem persists in (probably) all versions including the gcc in Ubuntu 16.04 and up to the one in Debian testing at the moment of this writing (4:6.2.1-1). The allocation modifier in question is specified here: http://pubs.opengroup.org/onlinepubs/9699919799/functions/fscanf.html Attached you can find an example program that shows the problem and explains it in more detail. -- Dipl.-Ing. Stefan Tauner Research and Development Embedded Systems Department University of Applied Sciences Technikum Wien Hoechstaedtplatz 6, 1200 Vienna, Austria T: +43 1 333 40 77-316 E: stefan.tau...@technikum-wien.at I: embsys.technikum-wien.at I: www.technikum-wien.at#include #include int main() { char *s, *p, *c; scanf("%ms %mc %m[a-z]", , , ); printf("s is %s, c is %c, p is %s\n", s, *c, p); free(s); free(c); free(p); } /* This produces... scanfm.c: In function âmainâ: scanfm.c:6:11: warning: ISO C does not support the 'm' scanf flag [-Wformat=] scanf("%ms %mc %m[a-z]", , , ); ^ scanfm.c:6:11: warning: ISO C does not support the 'm' scanf flag [-Wformat=] scanfm.c:6:11: warning: ISO C does not support the 'm' scanf flag [-Wformat=] etc. when compiled with GCC even if _POSIX_C_SOURCE is set to 200809L, e.g., gcc -Wall -D_POSIX_C_SOURCE=200809L -pedantic -std=c99 scanfm.c -o scanfm The 'm' allocation character is specified in POSIX 2008[1] and thus I think gcc should not warn if _POSIX_C_SOURCE or _XOPEN_SOURCE are set appropriately. 1: http://pubs.opengroup.org/onlinepubs/9699919799/functions/fscanf.html */
Bug#834957: flashrom: please make the build reproducible
On Sun, 21 Aug 2016 00:34:08 +0100 Chris Lamb <la...@debian.org> wrote: > Source: flashrom > Version: 0.9.9+r1954-1 > Severity: wishlist > Tags: patch > User: reproducible-bui...@lists.alioth.debian.org > Usertags: timestamps > X-Debbugs-Cc: reproducible-bui...@lists.alioth.debian.org > > Hi, > > Whilst working on the Reproducible Builds effort [0], I noticed > that flashrom could not be built reproducibly. Hi Chris, if I read the specs right then the date formatting should not happen at build time but at runtime to catch user locales. "Formatting MUST be deferred until runtime if an end user should observe the value in their own locale or timezone." That would make your patch not fully complying. I'll try to improve on that upstream for the next release. Thanks for the pointer. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#766461: gitg: upload 3.14 to unstable
Hello Dmitry, gitg users! I became aware of the complete rewrite and UI regressions only with the update of my Ubuntu boxes. It's amazing how far away from my daily needs this previously useful application has been pushed by the gnome mindset (or whatever). I assume I am not the only one who cannot stand the new interface... I deem the 3.x version a fork. It feels like Gnome 2 -> 3 all over again - just smaller :( Obviously I have not followed the development of gitg but I have been using it for over 5 years almost daily (at work and home). I do understand that the old 0.x branches wont be maintained by anybody in the future and that 3.x is already lurking in experimental. I am also quite aware that maintaining it with updated (gnome) libraries etc. would require quite some effort - at least in the beginning. I won't be able to dedicate enough time to do that on my own (and probably lack some gtk/glib experience as well) but I am willing to join any group working on this in case anybody is interested... I could not find any existing public efforts in this regard. Dmitry, I refrained from opening another bug report to spare you from having to close it as non-sense... no idea how such things should be handled properly due to their subjective nature so I just appended it to the best matching existing bug to make the above public. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#399076: Upstream
tags 399076 upstream This is still not fixed and contained in upstream Bash as well. I have reported it upstream. KR -- Dipl.-Ing. Stefan Tauner Research and Development Embedded Systems Department University of Applied Sciences Technikum Wien Hoechstaedtplatz 6, 1200 Vienna, Austria T: +43 1 333 40 77-316 E: stefan.tau...@technikum-wien.at I: embsys.technikum-wien.at I: www.technikum-wien.at
Bug#802342: flashrom: add arm64?
On Mon, 19 Oct 2015 18:38:53 +0100 Edmund Grimley Evans <edmund.grimley.ev...@gmail.com> wrote: > Source: flashrom > Version: 0.9.7+r1852-1.1 > Tags: patch > > I have no idea whether this package is useful on arm64, but it's easy > enough to build it with the attached trivial patch. Hi, I am the upstream maintainer. Yes, it is useful and we have added support upstream for aarch64 (as well as PPC64) in r1864. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner
Bug#775428: flashrom FTBFS on mipsel
Hi Uwe, can you please trigger building the 0.9.8-rc1 candidate (before I release the final in about a week)? -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#775428: flashrom FTBFS on mipsel
On Thu, 15 Jan 2015 16:12:23 + Dejan Latinovic dejan.latino...@imgtec.com wrote: Patch that conatins needed changes is attached. This chage is aviale upstream, along with other changes. With this patch I was able to build flashrom for mipsel locally. Could you please consider including these changes? Here is commit that conatins my fix. https://github.com/stefanct/flashrom/commit/138245ea51472211ec12e7b4d176591f3f18ce38 If needed, we could use all chages from this commit. Hello Dejan (and Hello Uwe too), I am planning to release a new (upstream) flashrom stable release (0.9.8) in the next weeks. This will include said patch as well as many other changes committed since the last snapshot was taken for Debian. It would probably make sense to merge that one to sid instead of backporting anything... -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#746734: flashrom: Please upgrade to 0.9.7 (+ patches)
Package: flashrom Severity: wishlist Dear Uwe, flashrom 0.9.7 was released on 2013-08-14. Since then an additional patch was pushed to the stable branch (r1752). We have tried to reach you multiple times since then on IRC to get the package updated but we got (almost?) no response. This report shall make this request more official and also persistent and verifiable, so that we can orphan the package later if need be. I would prefer if you could continue maintaining flashrom, but not updating without any rationale for such a long time is not really acceptable for a software like flashrom that has to cope with changes in hardware at a very low level. Kind regards, Stefan Tauner -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#690478: (no subject)
Control: retitle -1 flashrom does not probe as expected with no options Tags: upstream, upstream-fixed Hello, According to the docs/man page: If no operation is specified, flashrom will only probe for flash chips. It is recommended that if you try flashrom the first time on a system, you run it in probe-only mode and check the output. However when run with no options i get an error about no programmer parameter: … Hello Douglas, it does not say If no parameter is given, … but *operation*. I see your point though and I have added a hint regarding the -p option to the paragraph in question (upstream, not to the Debian package). It now reads: You can specify one of -h, -R, -L, -z, -E, -r, -w, -v or no operation. If no operation is specified, flashrom will only probe for flash chips. It is recommended that if you try flashrom the first time on a system, you run it in probe-only mode and check the output. Also you are advised to make a backup of your current ROM contents with -r before you try to write a new image. All operations involving any chip access (probe/read/write/...) require the -p/--programmer option to be used (please see below). I hope this is in your interest. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/debian-bugs-dist
Bug#637028: dmidecode does falsely use bit 7 when decoding the chassis-type
Package: dmidecode the bug is present in all vanilla versions since at least 2.9 and in various Debian-based distributions (Ubuntu, grml). the attached patch was also sent upstream (and to Ubuntu). for details please see the patch. -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner From 888158fc7e0c7b9bc33281c2f80c5501f01b304e Mon Sep 17 00:00:00 2001 From: Stefan Tauner stefan.tau...@student.tuwien.ac.at Date: Sun, 7 Aug 2011 20:24:44 +0200 Subject: [PATCH] Make dmi_chassis_type aware of the lock bit. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit To: dmidecode-de...@nongnu.org Previously all bits of the parameter passed to dmi_chassis_type were used to derive the chassis type although the 7th bit indicates a lock and only bits 6:0 encode the chassis type (7.4 System Enclosure or Chassis (Type 3), offset 05h). This is ok as long as the input is masked as it was done in dmi_decode, but it was forgotten in dmi_table_string, resulting in wrong output if there is a lock present: dmidecode -s chassis-type OUT OF SPEC although the normal output is correct: [â¦] Handle 0x0003, DMI type 3, 17 bytes Chassis Information Manufacturer: Chassis Manufacture Type: Desktop Lock: Present [â¦] dump (the 5th byte (83) is the interesting one): dmidecode -t chassis -u SMBIOS 2.3 present. Handle 0x0003, DMI type 3, 17 bytes Header and Data: 03 11 03 00 01 83 02 03 04 03 03 03 03 01 00 00 00 Tested with current CVS code on a Laptop without a lock (by me) and on the Desktop board dumped above (by Florian Zumbiehl, thanks!). --- dmidecode.c |3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/dmidecode.c b/dmidecode.c index f7b23c1..af2bfc5 100644 --- a/dmidecode.c +++ b/dmidecode.c @@ -532,6 +532,7 @@ static const char *dmi_chassis_type(u8 code) Blade Enclosing /* 0x1D */ }; + code = 0x7F; /* bits 6:0 are chassis type, 7th bit is the lock bit */ if (code = 0x01 code = 0x1D) return type[code - 0x01]; return out_of_spec; @@ -3237,7 +3238,7 @@ static void dmi_decode(const struct dmi_header *h, u16 ver) printf(\tManufacturer: %s\n, dmi_string(h, data[0x04])); printf(\tType: %s\n, -dmi_chassis_type(data[0x05] 0x7F)); +dmi_chassis_type(data[0x05])); printf(\tLock: %s\n, dmi_chassis_lock(data[0x05] 7)); printf(\tVersion: %s\n, -- 1.7.1
Bug#626119: New upstream version of avr-libc available (1.7.1)
Package: avr-libc Version: 1:1.6.8-2: all Hello Hakan (et al)! please consider updating avr-libc. It is quite outdated (1.6.7 is 22 months old) and misses support for a lot of new models; see: http://svn.savannah.nongnu.org/viewvc/tags/?root=avr-libc and http://www.nongnu.org/avr-libc/NEWS.txt Thank you! -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#573254: BTS#573254 rsnapshot shouldn't bail on absent backup points
hello the problem is that rsnapshot does fear the unknown (line 1004 in ubuntu's rsnapshot). below is the last else clause checking different formats of the backup source inside the sub parse_config_file: # fear the unknown } else { config_err($file_line_num, $line - Source directory \$src\ doesn't exist); next; } in some situations it should not fear, because it does not need (the unknown) source directory: - as mentioned by the op it should not fear if it is just rotating - when invoking diff or du this leads to situations where 'rsnapshot -c ... diff' with one config file bails but using another config file, where the source exists (but does not matter) _to diff directories of the same snapshot root_ works flawlessly: /data/backup/system# rsnapshot -c /root/backup-scripts/settings/hostname/rsnapshot-root.conf diff bootly.[01] rsnapshot encountered an error! The program was invoked with these options: /usr/bin/rsnapshot -c \ /root/backup-scripts/settings/hostname/rsnapshot-root.conf diff bootly.0 \ bootly.1 ERROR: /root/backup-scripts/settings/hostname/rsnapshot-root.conf on line 223: ERROR: backup /mnt/root-snapshot root/ \ exclude=dev,exclude=tmp,exclude=lib/udev/,exclude=var/spool \ - Source directory /mnt/root-snapshot doesn't exist ERROR: - ERROR: Errors were found in /root/backup-scripts/settings/hostname/rsnapshot-root.conf, ERROR: rsnapshot can not continue. If you think an entry looks right, make ERROR: sure you don't have spaces where only tabs should be. the second one uses the default config file (which i don't use normally) with this backup points which all exist: backup /home/ localhost/ backup /etc/ localhost/ backup /usr/local/ localhost/ /data/backup/system# rsnapshot diff bootly.[01] Comparing bootly.1 to bootly.0 Between bootly.1 and bootly.0: 92 were added, taking 6362802 bytes; 85 were removed, saving 6292118 my perl skills are non-existent and i did not read a lot of rsnapshot's code, but my general idea to fix this problem would be: - define all cases where the source directory is not need - set a global variable or argument of parse_config_file to indicate when this is the case - skip the source directory check in parse_config_file according to this information -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#611784: geteltorito in genisoimage doesn't extract hard drive images properly
Package: genisoimage Source: cdrkit Version: 9:1.1.11-1 Version: 9:1.1.10-1ubuntu3 geteltorito is outdated in Debian and Ubuntu. Please update. I'm quoting the Ubuntu bug report[1] below: geteltorito fails to extract hard drive images from el torito boot cd's. The example I have currently is extracting Thinkpad BIOS update images from the boot CDs available from Lenovo. This has been fixed in a newer version of the geteltorito script. The updated script can be found at http://www.uni-koblenz.de/~krienke/ftp/noarch/geteltorito/geteltorito 1: http://bugs.launchpad.net/ubuntu/+source/cdrkit/+bug/530530 -- Kind regards/Mit freundlichen Grüßen, Stefan Tauner -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org