Re: [DNG] Bootloaders (was: SystemD's brownie points over non-systemd OSs)
Peter Olson wrote: > I have a machine in that state right now, and rather than try to debug it at > the Grub prompt, I am just going to reinstall the system. That's a bit like the old "I'm buying a new car because the ashtray is full" joke. If you've managed to screw up your kernel and/or init image then that's not a grub problem - a bit like blaming Goodyear (or whoever) because you drove over a rock and shredded both the tyre and rim. If you haven't then the system is recoverable. Many installers have a rescue option (may be under an advanced submenu) that will allow you to boot the system and mount your chosen partition as a temporary root. Then "update-grub" "grub-install /dev/..." should fix it. Or you can do it manually. Boot off a live disk and mount /dev/... /mnt mount /dev/... /mnt/boot (if you use a boot filesystem) mount -o bind /proc /mnt/proc mount -o bind /sys /mnt/sys mount -o bind /dev /mnt/dev chroot /mnt - fix your grub setup exit the chroot,, unmount filesystems, sync, reboot That usually does it for me. Or simpler than the above, give http://www.supergrubdisk.org a try. It was suggested to me last week, and it's really good. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
Hi Richard, On 08/10/2016 07:15 AM, richard lucassen wrote: The /urs/etc dir is for the example (documentation) about how to configure it. I've never seen any /usr/etc, but /usr/local/etc already exists. I respected all the directories established by Jude Nelson, changing only the destdir "/usr/local" by "/usr" in the Makefile, via *quilt*. This is a basic rule. Thanks a lot:) Aitor. Forget these words. I said a nonsense. I think that /usr is only for the libudev-compat package. The real location for vdevd is /sbin. So, you are right. Today i will work on that. I'll also try to boot vdev in an initramfs. I'll change the Makefile: .PHONY: all all: $(MAKE) -C vdevd PREFIX= $(MAKE) -C libudev-compat PREFIX= INCLUDE_PREFIX=/usr $(MAKE) -C fs PREFIX= $(MAKE) -C example PREFIX= $(MAKE) -C hwdb PREFIX= .PHONY: install install: $(MAKE) -C vdevd install PREFIX= $(MAKE) -C libudev-compat install PREFIX= INCLUDE_PREFIX=/usr $(MAKE) -C fs install PREFIX= $(MAKE) -C example install PREFIX= $(MAKE) -C hwdb install PREFIX= [...] I also will add: update-rc.d udev disable update-rc.d udev-finish disable to the *postinst* file. Cheers, Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
I've now made a snapshot of the vdev files from the working disk. available at www.realthing.com.au/files/vdev/vdev-snapshot.tgz. As I mentioned before, it came down to a couple changes to the config file and the udev-compat.sh helper (maybe not needed), then adding dameonlet and the acls directory with its thing, and setting a handful of links. The more major thing was the addition of the initramfs making (in the root director of the snapshot), which I took from github, and editing it to deal with the /usr prefix. I made it a snapshot for forensic study, and you can choose how to roll it into the packages. Note that I added "loop" to modules because the hardware database is a squasfs; I haven't actually verified that it's really needed but just left it so. Ralph. On 11/08/16 17:19, aitor_czr wrote: Hi Richard, On 08/10/2016 07:15 AM, richard lucassen wrote: The /urs/etc dir is for the example (documentation) about how to configure it. I've never seen any /usr/etc, but /usr/local/etc already exists. I respected all the directories established by Jude Nelson, changing only the destdir "/usr/local" by "/usr" in the Makefile, via *quilt*. This is a basic rule. Thanks a lot:) Aitor. Forget these words. I said a nonsense. I think that /usr is only for the libudev-compat package. The real location for vdevd is /sbin. So, you are right. Today i will work on that. I'll also try to boot vdev in an initramfs. I'll change the Makefile: .PHONY: all all: $(MAKE) -C vdevd PREFIX= $(MAKE) -C libudev-compat PREFIX= INCLUDE_PREFIX=/usr $(MAKE) -C fs PREFIX= $(MAKE) -C example PREFIX= $(MAKE) -C hwdb PREFIX= .PHONY: install install: $(MAKE) -C vdevd install PREFIX= $(MAKE) -C libudev-compat install PREFIX= INCLUDE_PREFIX=/usr $(MAKE) -C fs install PREFIX= $(MAKE) -C example install PREFIX= $(MAKE) -C hwdb install PREFIX= [...] I also will add: update-rc.d udev disable update-rc.d udev-finish disable to the *postinst* file. Cheers, Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On 08/10/2016 10:46 AM, Ralph Ronnquist wrote: I've now made a snapshot of the vdev files from the working disk. available at www.realthing.com.au/files/vdev/vdev-snapshot.tgz. As I mentioned before, it came down to a couple changes to the config file and the udev-compat.sh helper (maybe not needed), then adding dameonlet and the acls directory with its thing, and setting a handful of links. The more major thing was the addition of the initramfs making (in the root director of the snapshot), which I took from github, and editing it to deal with the /usr prefix. I made it a snapshot for forensic study, and you can choose how to roll it into the packages. Note that I added "loop" to modules because the hardware database is a squasfs; I haven't actually verified that it's really needed but just left it so. Ralph. Thanks ! Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
[DNG] Your computer clock: Was:Re: vdev
On Thu, 2016-08-11 at 02:00 +0200, aitor_czr wrote: > > Aitor. > Aitor, your computer clock is running ahead again. Can you please sync with ntpdate regularly in a cron script (or replace the CMOS battery?) Thanks! ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Your computer clock: Was:Re: vdev
Hi Svante, On 08/10/2016 11:16 AM, Svante Signell wrote: On Thu, 2016-08-11 at 02:00 +0200, aitor_czr wrote: > > Aitor. > Aitor, your computer clock is running ahead again. Can you please sync with ntpdate regularly in a cron script (or replace the CMOS battery?) Thanks! My clock is right: aitor@gnuinos:~$ date Thu Aug 11 11:14:02 CEST 2016 It only happens in another computer. My second handed Toshiba works fine :) Cheers, Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
Hi Ralph, On 08/10/2016 10:46 AM, Ralph Ronnquist wrote: Note that I added "loop" to modules because the hardware database is a squasfs; I haven't actually verified that it's really needed but just left it so. Ralph. All the required modules are included in the kernel: SQUASHFS LOOP SD_MOD Cheers, Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
Hello. Thanks to a friendly help, I've found a few mails and articles which deserve to be read: Udev on non-systemd is a dead-end: https://lists.freedesktop.org/archives/systemd-devel/2012-August/006066.html To face issues with Udev, the Linux kernel team has started implementing firmware loading by the kernel itself, by 2012: http://lwn.net/Articles/518942/ They also consider loading the modules directly, in both cases ressorting to a hotplugger only in case of failure: http://marc.info/?l=linux-kernel&m=134921751125528&w=4 Didier ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Your computer clock: Was:Re: vdev
aitor_czr wrote: > My clock is right: > > aitor@gnuinos:~$ date > Thu Aug 11 11:14:02 CEST 2016 Err, no it isn't - unless you've found the secret of time travel ! You're a day ahead of us. Your clock says 11th Aug, in the rest of the world it's still the 10th Aug. And from your message headers : > Received: from [*.*.*.*] (.euskaltel.es [62.99.112.114]) (Authenticated > sender: ***@***) by player726.ha.ovh.net (Postfix) with ESMTPSA id > C7C5D2A0075; Wed, 10 Aug 2016 11:21:38 +0200 (CEST) ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On Thu, 11 Aug 2016 11:23:03 +0200 aitor_czr wrote: > On 08/10/2016 10:46 AM, Ralph Ronnquist wrote: > > Note that I added "loop" to modules because the hardware database > > is a squasfs; I haven't actually verified that it's really needed > > but just left it so. > > All the required modules are included in the kernel: > > SQUASHFS > LOOP > SD_MOD I suppose he means to add the modules to /etc/modules to be sure they are loaded at boot time. # cat /etc/modules loop sd_mod squashfs R. -- richard lucassen http://contact.xaq.nl/ ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On Tue, Aug 09, 2016 at 09:14:09PM +0200, richard lucassen wrote: > Oh, BTW, when packaging vdevd, I'd opt for the /etc/vdev/ dir and not > the /usr/etc/vdev/ dir. All these files are supposed to be conffiles, and thus belong under /etc. Systemd moved them to /usr because Red Hat has abysmal configuration handling, and, unlike Debian, has no real tools to update conffiles possibly changed by the user on upgrades. Thus Red Hat puts everything in /usr and expects you to copy files to /etc to override -- you don't get any notification whatsoever that your config became outdated and needs to be adjusted to let the program in question work. It's a "must" requirement of the Policy (10.7.2), thus systemd-udev would have a RC bug if systemd wasn't above the law. While derivatives may have their own policies and thus are not bound by Debian's, vdev might become necessary when udev goes systemd-only (if vdev won't yet be in shape at that time, eudev is plan B), thus it'd be nice for its packaging to have the conffiles in /etc. Meow! -- An imaginary friend squared is a real enemy. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On Wed, 10 Aug 2016 18:46:36 +1000 Ralph Ronnquist wrote: > I've now made a snapshot of the vdev files from the working disk. > available at www.realthing.com.au/files/vdev/vdev-snapshot.tgz. logfile=/run/vdev/vdevd.log I'd rather choose logfile=/var/log/vdev/vdevd.log as /run/ is a ramdisk ;-) > As I mentioned before, it came down to a couple changes to the config > file and the udev-compat.sh helper (maybe not needed), then adding > dameonlet and the acls directory with its thing, and setting a > handful of links. You probably have no list of the changes you made? > The more major thing was the addition of the initramfs making (in the > root director of the snapshot), which I took from github, and editing > it to deal with the /usr prefix. > > I made it a snapshot for forensic study, and you can choose how to > roll it into the packages. Note that I added "loop" to modules > because the hardware database is a squasfs; I haven't actually > verified that it's really needed but just left it so. I installed vdev like in INSTALL.md, but it seems to have problems. vdev starts, but quits. It creates most devices, but does not set the apropiate ownerships and permissions. When I start vdev manually, it starts but refuses to set the apropiate ownerships and permissions. But when running: /sbin/vdevd -v2 -c /etc/vdev/vdevd.conf -l /var/log/vdev/vdev.log /dev everything seems to work well, the apropiate ownerships and permissions are set to the expected settings. IOW: when run manually it's ok. It might have something to do with initramfs which refuses to compile, due to an old bug. The workaroud is to: ln -s /dev/sda5 /805 in order to make it work again. But generating the initramfs from the example directory stops with an error and an empty initramfs file. The thing I did was symlinking the libudev.so.1 ls -al /lib/i386-linux-gnu/libudev.so.1 lrwxrwxrwx 1 root root 17 Aug 10 14:08 /lib/i386-linux-gnu/libudev.so.1 -> /lib/libudev.so.1 and to run a update-initramfs -u -v Anyone a hint? R. -- richard lucassen http://contact.xaq.nl/ ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On Wed, 10 Aug 2016 14:22:07 +0200 richard lucassen wrote: > But when running: > > /sbin/vdevd -v2 -c /etc/vdev/vdevd.conf -l /var/log/vdev/vdev.log /dev > > everything seems to work well, the apropiate ownerships and > permissions are set to the expected settings. Ownerships and permissions are set when starting, but when adding a usb stick, it sets the ownerships and permissions to root.root/600 :-( There are still some glitches here and there I fear. Anyway, grosso modo it seems to work, although it still needs soms manual tweaking :) -- richard lucassen http://contact.xaq.nl/ ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Your computer clock: Was:Re: vdev
On Wed, Aug 10, 2016 at 10:59:07AM +0100, Simon Hobson wrote: > aitor_czr wrote: > > > My clock is right: > > > > aitor@gnuinos:~$ date > > Thu Aug 11 11:14:02 CEST 2016 > > Err, no it isn't - unless you've found the secret of time travel ! You're a > day ahead of us. > > Your clock says 11th Aug, in the rest of the world it's still the 10th Aug. > And from your message headers : Might he live just west of the International Date Line? -- hendrik > > > Received: from [*.*.*.*] (.euskaltel.es [62.99.112.114]) (Authenticated > > sender: ***@***) by player726.ha.ovh.net (Postfix) with ESMTPSA id > > C7C5D2A0075; Wed, 10 Aug 2016 11:21:38 +0200 (CEST) > > > ___ > Dng mailing list > Dng@lists.dyne.org > https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On 10/08/16 22:22, richard lucassen wrote: On Wed, 10 Aug 2016 18:46:36 +1000 Ralph Ronnquist wrote: I've now made a snapshot of the vdev files from the working disk. available at www.realthing.com.au/files/vdev/vdev-snapshot.tgz. logfile=/run/vdev/vdevd.log I'd rather choose logfile=/var/log/vdev/vdevd.log as /run/ is a ramdisk ;-) So would I :-) but I got the impression /var (or /) is read-only when vdevd starts, which made it spit the dummy. That's why I tried with directing it to /run. I guess the logging needs som hands-on As I mentioned before, it came down to a couple changes to the config file and the udev-compat.sh helper (maybe not needed), then adding dameonlet and the acls directory with its thing, and setting a handful of links. You probably have no list of the changes you made? For the actually changed files, I have the context diff: diff -rc clean/usr/etc/vdev/vdevd.conf fixed/usr/etc/vdev/vdevd.conf *** clean/usr/etc/vdev/vdevd.conf 2016-07-02 04:09:23.0 +1000 --- fixed/usr/etc/vdev/vdevd.conf 2016-08-10 00:10:07.0 +1000 *** *** 1,12 [vdev-config] ! firmware=/usr/lib/firmware acls=/usr/etc/vdev/acls actions=/usr/etc/vdev/actions helpers=/usr/lib/vdev hwdb=/usr/lib/vdev/hwdb/hwdb.squashfs ifnames=/usr/etc/vdev/ifnames.conf ! pidfile=/usr/run/vdev/vdevd.pid default_permissions=0600 loglevel=debug ! logfile=/usr/var/log/vdev/vdevd.log preseed=/usr/lib/vdev/dev-setup.sh --- 1,13 [vdev-config] ! firmware=/lib/firmware acls=/usr/etc/vdev/acls actions=/usr/etc/vdev/actions helpers=/usr/lib/vdev hwdb=/usr/lib/vdev/hwdb/hwdb.squashfs + hwdb_loop=loop1 ifnames=/usr/etc/vdev/ifnames.conf ! pidfile=/run/vdev/vdevd.pid default_permissions=0600 loglevel=debug ! logfile=/run/vdev/vdevd.log preseed=/usr/lib/vdev/dev-setup.sh diff -rc clean/usr/lib/vdev/udev-compat.sh fixed/usr/lib/vdev/udev-compat.sh *** clean/usr/lib/vdev/udev-compat.sh 2016-07-02 04:08:31.0 +1000 --- fixed/usr/lib/vdev/udev-compat.sh 2016-08-10 17:55:56.007870574 +1000 *** *** 571,580 $(udev_event_generate_text "$VDEV_ACTION" "$VDEV_OS_DEVPATH" "$VDEV_OS_SUBSYSTEM" "$VDEV_OS_SEQNUM" "$VDEV_METADATA") EOF echo "event-put $_DEVICE_ID" >> /tmp/udev-compat.log echo "" >> /tmp/udev-compat.log -_RC=$? return $_RC } --- 571,580 $(udev_event_generate_text "$VDEV_ACTION" "$VDEV_OS_DEVPATH" "$VDEV_OS_SUBSYSTEM" "$VDEV_OS_SEQNUM" "$VDEV_METADATA") EOF +_RC=$? echo "event-put $_DEVICE_ID" >> /tmp/udev-compat.log echo "" >> /tmp/udev-compat.log return $_RC } - ends here The more major thing was the addition of the initramfs making (in the root director of the snapshot), which I took from github, and editing it to deal with the /usr prefix. I made it a snapshot for forensic study, and you can choose how to roll it into the packages. Note that I added "loop" to modules because the hardware database is a squasfs; I haven't actually verified that it's really needed but just left it so. I installed vdev like in INSTALL.md, but it seems to have problems. Just to be clear: you are now going back to the github source. My snapshot is from the new exercise against the gnuinos. (Except that I brought in the making of the initramfs, which was missing in the deb) vdev starts, but quits. I think that was the effect of trying to log on /var. It creates most devices, but does not set the apropiate ownerships and permissions. When I start vdev manually, it starts but refuses to set the apropiate ownerships and permissions. But when running: /sbin/vdevd -v2 -c /etc/vdev/vdevd.conf -l /var/log/vdev/vdev.log /dev everything seems to work well, the apropiate ownerships and permissions are set to the expected settings. IOW: when run manually it's ok. It might have something to do with initramfs which refuses to compile, due to an old bug. The workaroud is to: ln -s /dev/sda5 /805 in order to make it work again. But generating the initramfs from the example directory stops with an error and an empty initramfs file. The thing I did was symlinking the libudev.so.1 ls -al /lib/i386-linux-gnu/libudev.so.1 lrwxrwxrwx 1 root root 17 Aug 10 14:08 /lib/i386-linux-gnu/libudev.so.1 -> /lib/libudev.so.1 and to run a update-initramfs -u -v Anyone a hint? R. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Your computer clock: Was:Re: vdev
... about 16 hours East of East Australia; roughly in time zone +26 I think ;-0 On 10/08/16 23:14, Hendrik Boom wrote: On Wed, Aug 10, 2016 at 10:59:07AM +0100, Simon Hobson wrote: aitor_czr wrote: My clock is right: aitor@gnuinos:~$ date Thu Aug 11 11:14:02 CEST 2016 Err, no it isn't - unless you've found the secret of time travel ! You're a day ahead of us. Your clock says 11th Aug, in the rest of the world it's still the 10th Aug. And from your message headers : Might he live just west of the International Date Line? -- hendrik Received: from [*.*.*.*] (.euskaltel.es [62.99.112.114]) (Authenticated sender: ***@***) by player726.ha.ovh.net (Postfix) with ESMTPSA id C7C5D2A0075; Wed, 10 Aug 2016 11:21:38 +0200 (CEST) ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Your computer clock: Was:Re: vdev
On Wed, Aug 10, 2016 at 09:14:13AM -0400, Hendrik Boom wrote: > On Wed, Aug 10, 2016 at 10:59:07AM +0100, Simon Hobson wrote: > > aitor_czr wrote: > > > > > My clock is right: > > > > > > aitor@gnuinos:~$ date > > > Thu Aug 11 11:14:02 CEST 2016 > > > > Err, no it isn't - unless you've found the secret of time travel ! You're a > > day ahead of us. > > > > Your clock says 11th Aug, in the rest of the world it's still the 10th Aug. > > And from your message headers : > > Might he live just west of the International Date Line? 1. CEST gives you the time zone. 2. Time zones go only from -12:00 to +14:00, ie, no more than 12 hours ahead of the receiver (who also uses CEST). -- An imaginary friend squared is a real enemy. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On Wed, 10 Aug 2016 09:34:12 -0400 fsmithred wrote: > No SD_MOD here... > > $ grep SD_MOD /boot/config-3.16.0-4-amd64 find /lib/modules/ -name sd_mod.ko -- richard lucassen http://contact.xaq.nl/ ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On 08/11/2016 05:23 AM, aitor_czr wrote: > > Hi Ralph, > > On 08/10/2016 10:46 AM, Ralph Ronnquist wrote: >> Note that I added "loop" to modules because the hardware database is a >> squasfs; I haven't actually verified that it's really needed but just >> left it so. >> >> Ralph. > > All the required modules are included in the kernel: > > SQUASHFS > LOOP > SD_MOD > > Cheers, > > Aitor. > No SD_MOD here... $ grep SD_MOD /boot/config-3.16.0-4-amd64 $ -fsr ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Your computer clock: Was:Re: vdev
Hi Simon, On 08/10/2016 02:00 PM, Simon Hobson wrote: aitor_czr wrote: >My clock is right: > >aitor@gnuinos:~$ date >Thu Aug 11 11:14:02 CEST 2016 Err, no it isn't - unless you've found the secret of time travel ! You're a day ahead of us. Your clock says 11th Aug, in the rest of the world it's still the 10th Aug. And from your message headers : >Received: from [*.*.*.*] (.euskaltel.es [62.99.112.114]) (Authenticated sender: ***@***) by player726.ha.ovh.net (Postfix) with ESMTPSA id C7C5D2A0075; Wed, 10 Aug 2016 11:21:38 +0200 (CEST) I'm not Steven Spielberg :) Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On Wed, 10 Aug 2016 23:21:36 +1000 Ralph Ronnquist wrote: > > logfile=/run/vdev/vdevd.log > > > > I'd rather choose > > > > logfile=/var/log/vdev/vdevd.log > > > > as /run/ is a ramdisk ;-) > > So would I :-) but I got the impression /var (or /) is read-only when > vdevd starts, which made it spit the dummy. That's why I tried with > directing it to /run. I guess the logging needs som hands-on Que? Even /tmp/ does not work. Indeed, using /run/ makes vdevd start. Is this a feature or a bug? :) BTW: Does your vdevd set the right permissions? I will have al look at your diffs in 5 minutes -- richard lucassen http://contact.xaq.nl/ ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
Hi fsmithred, On 08/10/2016 03:36 PM, fsmithred wrote: On 08/11/2016 05:23 AM, aitor_czr wrote: > >Hi Ralph, > >On 08/10/2016 10:46 AM, Ralph Ronnquist wrote: >>Note that I added "loop" to modules because the hardware database is a >>squasfs; I haven't actually verified that it's really needed but just >>left it so. >> >>Ralph. > >All the required modules are included in the kernel: > >SQUASHFS >LOOP >SD_MOD > >Cheers, > > Aitor. > No SD_MOD here... $ grep SD_MOD /boot/config-3.16.0-4-amd64 $ -fsr SCSI, maybe? There are alias... Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Timezones: Was: Your computer clock: Was:Re: vdev
Hendrik Boom wrote: > On Wed, Aug 10, 2016 at 10:59:07AM +0100, Simon Hobson wrote: >> Err, no it isn't - unless you've found the secret of time travel ! You're a >> day ahead of us. >> >> Your clock says 11th Aug, in the rest of the world it's still the 10th Aug. >> And from your message headers : > > Might he live just west of the International Date Line? Two reasons why that wouldn't explain it. Firstly, he gives the time as CEST (UTC+2h) Secondly, at the time I wrote the email, which was some time after he'd sent his, he'd be just coming up to midnight if he was in that very small list of places in the UTC+14 timezone - eg Kiritimati, Christmas Island, Kiribati, Samoa with DST http://www.timeanddate.com/worldclock/kiribati/kiritimati So while I was at the time unaware (or had forgotten) about those select UTC+14 places, I was correct in saying that the rest of the world was still on 10th and that would still be correct for another 53 seconds after I sent the message :-) Incidentally, the stories behind some of these bits of trivia can be fascinating. I vaguely recall some islands shifting by a day - ie declaring them to now be in a UTC+nn timezone instead of UTC-nn. I suspect the "we are first to see in the new year" (and hence get a bit of tourist trade) factor may have been involved, but I vaguely recall it was more a case of them having more affinity with the Russia/China/Australia part of the world (and all the islands in that expanse of ocean) than with America - and thus it making sense to be in the same day as those (to us) eastern countries. Ah, I stand corrected, according to https://en.wikipedia.org/wiki/UTC%2B14:00 it was because parts of the colony were on different sides of the IDL and that meant that they only had 4 working days in common - Monday on one island would be Sunday on another, and similarly with Friday/Saturday (in the reverse direction). For good measure, Australia has 1/2 hour offsets in some of it's time zones (and even one with a 1/4 hour offset !) - which I would have thought must cause "a certain amount of confusion" http://www.timeanddate.com/time/australia/time-zones-background.html I guess some of these oddities had their reasons at the time - the UK used to have local times which varied across the country, it was only the expansion of the railways that made a common timezone "useful". And since then, there's probably a "that's our history, you can't change it" response to any suggestions of rationalisation ! As I say, can be fascinating - and a welcome diversion from work ;-) ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On Wed, 10 Aug 2016 23:21:36 +1000 Ralph Ronnquist wrote: > > You probably have no list of the changes you made? > > For the actually changed files, I have the context diff: Mine: # cat /etc/vdev/vdevd.conf [vdev-config] firmware=/lib/firmware acls=/etc/vdev/acls actions=/etc/vdev/actions helpers=/lib/vdev hwdb=/lib/vdev/hwdb/hwdb.squashfs ifnames=/etc/vdev/ifnames.conf hwdb_loop=loop1 pidfile=/run/vdevd.pid default_permissions=0600 loglevel=debug logfile=/run/vdev.log preseed=/lib/vdev/dev-setup.sh (all files exist) > diff -rc clean/usr/lib/vdev/udev-compat.sh > echo "event-put $_DEVICE_ID" >> /tmp/udev-compat.log > echo "" >> /tmp/udev-compat.log > > -_RC=$? What are the consequences of moving this _RC? > +_RC=$? > echo "event-put $_DEVICE_ID" >> /tmp/udev-compat.log > echo "" >> /tmp/udev-compat.log After a boot, most permissions are root.root/600. When restarting vdev, the permissions are set correctly. When adding a usb stick, the permissions of sdb are set to the defaults root.root/600. After restarting vdev the permissions are set to the correct values: root.disk and 660. So there must be some or other glitch that jumps the permission settings. R. -- richard lucassen http://contact.xaq.nl/ ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
[DNG] Security news about TCP weakness
For those of you so inclined. Is this important, old news or just academic posturing? golinux -- https://www.sciencedaily.com/releases/2016/08/160809143253.htm Serious security threat to many Internet users highlighted Communications involving Linux and Android systems can be compromised quickly, easily and from anywhere Date:August 9, 2016 Source:University of California - Riverside Summary:Researchers have identified a weakness in the Transmission Control Protocol of all Linux operating systems since late 2012 that enables attackers to hijack users' internet communications completely remotely. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
On 08/10/2016 04:26 AM, Didier Kryn wrote: Hello. Thanks to a friendly help, I've found a few mails and articles which deserve to be read: Udev on non-systemd is a dead-end: So.. then.. basically any Linux distro which uses udev to populate /dev/ is going to be S.O.L? Including Slackware presumably? ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
[DNG] read-only filesystem using NETINST inside VirtualBox
Hi Devuanites, I'm trying to netinstall inside a VirtualBox VM and the root fs gets mounted read-only while extracting the base system: https://postimg.org/image/j8l7vp6uj/ https://postimg.org/image/hiwepyw6d/ https://postimg.org/image/pkk3j9q6d/ This is where it all starts, unpacking util-linux: https://postimg.org/image/dd2b549q3/ It tried it two consecutive times with same results :( The virtual machine's disk is SATA. Can anybody give me some hint on this? Thanks, Emiliano. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
dev wrote: >>Udev on non-systemd is a dead-end: > > So.. then.. basically any Linux distro which uses udev to populate /dev/ is > going to be S.O.L? Including Slackware presumably? That's about it - and I suspect that Poettering "isn't upset" by that. But reading the original links, he is clearly saying "I'll break stuff whenever *I* think it's right and I don't care how much work it makes for others in fixing the result". Proving (yet again) that he is not fit to be "in charge" of a critical bit of infrastructure. Reading the other threads from the last day or two ... How long before he decides that Grub needs "improving" ? ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] read-only filesystem using NETINST inside VirtualBox
Sorry,
This is VirtualBox's fault:
00:16:08.595972 AioMgr0-N: Request 0x0814cbd900 failed with
rc=VERR_TRY_AGAIN, migrating endpoint /.vmdk to failsafe manager.
00:16:08.890843 AIOMgr: I/O manager 0x08212fa4a0 encountered a critical
error (rc=VERR_FILE_AIO_NO_REQUEST) during operation. Falling back to
failsafe mode. Expect reduced performance
00:16:08.914567 AIOMgr: Error happened in
/wrkdirs/usr/ports/emulators/virtualbox-ose/work/VirtualBox-4.3.38/src/VBox/VMM/VMMR3/PDMAsyncCompletionFileNormal.cpp:(1664){int
pdmacFileAioMgrNormal(RTTHREAD, void *)}
https://forums.virtualbox.org/viewtopic.php?f=3&t=46722
On Wed, Aug 10, 2016 at 12:56 PM, Emiliano Marini <
emilianomarin...@gmail.com> wrote:
> Hi Devuanites,
>
> I'm trying to netinstall inside a VirtualBox VM and the root fs gets
> mounted read-only while extracting the base system:
>
> https://postimg.org/image/j8l7vp6uj/
>
> https://postimg.org/image/hiwepyw6d/
>
> https://postimg.org/image/pkk3j9q6d/
>
> This is where it all starts, unpacking util-linux:
>
> https://postimg.org/image/dd2b549q3/
>
> It tried it two consecutive times with same results :(
>
> The virtual machine's disk is SATA.
>
> Can anybody give me some hint on this?
>
> Thanks,
> Emiliano.
>
___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
dev writes: > On 08/10/2016 04:26 AM, Didier Kryn wrote: >> Hello. Thanks to a friendly help, I've found a few mails and >> articles which deserve to be read: >> >> Udev on non-systemd is a dead-end: > > So.. then.. basically any Linux distro which uses udev to populate > /dev/ is going to be S.O.L? Including Slackware presumably? This is seriously old news (dating back to 2012). udev became mandatory on Linux by eating hotplug. systemd is supposed to become mandatory on Linux by eating udev. There are two solutions to this problem: 1) Stop using udev. 2) Fork udev. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
I wrote: > But reading the original links, he is clearly saying "I'll break stuff > whenever *I* think it's right and I don't care how much work it makes for > others in fixing the result". However ... It does sound like this was an area potentially in want of some looking at. However, the way to do that would have been to discuss it rationally with the kernel guys and agree a way forward - putting the fix in well before the breakage. "Co-operation ? No, never heard of it" seems to be their motto. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On 08/11/2016 09:45 AM, aitor_czr wrote: > > Hi fsmithred, > > On 08/10/2016 03:36 PM, fsmithred wrote: >> On 08/11/2016 05:23 AM, aitor_czr wrote: >>> > >>> >Hi Ralph, >>> > >>> >On 08/10/2016 10:46 AM, Ralph Ronnquist wrote: >>Note that I added "loop" to modules because the hardware database is a >>squasfs; I haven't actually verified that it's really needed but just >>left it so. >> >>Ralph. >>> > >>> >All the required modules are included in the kernel: >>> > >>> >SQUASHFS >>> >LOOP >>> >SD_MOD >>> > >>> >Cheers, >>> > >>> > Aitor. >>> > >> No SD_MOD here... >> >> $ grep SD_MOD /boot/config-3.16.0-4-amd64 >> $ >> >> >> -fsr > > SCSI, maybe? > > There are alias... > >Aitor. > > > Yes, lots of SCSI. This one? CONFIG_SCSI_MOD=m -fsr ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
On Wed, Aug 10, 2016 at 10:51:22AM -0500, dev wrote: > > > On 08/10/2016 04:26 AM, Didier Kryn wrote: > > Hello. Thanks to a friendly help, I've found a few mails and > > articles which deserve to be read: > > > > Udev on non-systemd is a dead-end: > > So.. then.. basically any Linux distro which uses udev to populate /dev/ is > going to be S.O.L? Including Slackware presumably? Past tense would be more precise. Current udev _cannot_ be used to populate /dev, it doesn't contain any mknod() calls enymore. -- Tomasz Torcz"Funeral in the morning, IDE hacking xmpp: zdzich...@chrome.plin the afternoon and evening." - Alan Cox ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Security news about TCP weakness
Go Linux wrote: > For those of you so inclined. Is this important, old news or just academic > posturing? I think it's all three ! It looks very much related to a CVE from 2004 https://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2004-0230 Fundamentally, if someone can spoof a packet with the right source IP (trivial), right source port (may or may not be trivial), and (most importantly) the right sequence number - they can pretend to be the other end in any TCP connection. That's the basic issue, and there really isn't any way around that - if you receive a packet with the right headers then it's indistinguishable from any other packet with the right headers. Address is trivial to spoof - you only need to know that there is a conversation going on and the addresses of the end points. Port numbers are bit harder - one end is likely to be a well known port (eg port 80 for HTTP), but the other (the originating end) is likely to be a random number between 1024 and 65535, thus making it less than trivial to guess. The sequence number is (AIUI, fundamentally) there to allow packet identification within the stream so the data stream can be re-assembled with all the bits in the right order - and for this function can simply be a counter. Because a simple counter makes this sort of attack less difficult, I think things got changed so it's a pseudo-random sequence - as long as both ends know the rules, the sequence can be anything. The simplest thing you can do is to send things like RST packets and terminate the connection - thus causing a denial of service. With a bit more work, you can in theory inject false data packets and if you get them in, in the right time, they will be used instead of the real packet sent by the real other end. I can't help thinking that without being able to see any traffic, just figuring out which source ports are in use, AND the exact timing of the communication, is going to be sufficiently non-trivial as to make "I can alter contents of [web pages | emails | whatever ]" claims somewhat suspect. So I suspect that this new (if it is) revelation doesn't really make for a massive new attack vector. The key thing is figuring out what sequence number the other end will be sending, and the paper claims to have figured out a new and better way of doing that. I really don't know enough about the subject to know whether this paper is news or, as you suggest, just academic posturing. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
On Wed, 10 Aug 2016 16:59:58 +0100 Simon Hobson wrote: > How long before he decides that Grub needs "improving" ? How long before he decides that the kernel needs "improving"? -- richard lucassen http://contact.xaq.nl/ ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] read-only filesystem using NETINST inside VirtualBox
Solved enabling "Host I/O Cache" in the SATA Controller attributes.
On Wed, Aug 10, 2016 at 1:06 PM, Emiliano Marini wrote:
> Sorry,
>
> This is VirtualBox's fault:
>
> 00:16:08.595972 AioMgr0-N: Request 0x0814cbd900 failed with
> rc=VERR_TRY_AGAIN, migrating endpoint /.vmdk to failsafe manager.
> 00:16:08.890843 AIOMgr: I/O manager 0x08212fa4a0 encountered a
> critical error (rc=VERR_FILE_AIO_NO_REQUEST) during operation. Falling back
> to failsafe mode. Expect reduced performance
> 00:16:08.914567 AIOMgr: Error happened in /wrkdirs/usr/ports/emulators/
> virtualbox-ose/work/VirtualBox-4.3.38/src/VBox/VMM/VMMR3/
> PDMAsyncCompletionFileNormal.cpp:(1664){int pdmacFileAioMgrNormal(RTTHREAD,
> void *)}
>
> https://forums.virtualbox.org/viewtopic.php?f=3&t=46722
>
>
> On Wed, Aug 10, 2016 at 12:56 PM, Emiliano Marini <
> emilianomarin...@gmail.com> wrote:
>
>> Hi Devuanites,
>>
>> I'm trying to netinstall inside a VirtualBox VM and the root fs gets
>> mounted read-only while extracting the base system:
>>
>> https://postimg.org/image/j8l7vp6uj/
>>
>> https://postimg.org/image/hiwepyw6d/
>>
>> https://postimg.org/image/pkk3j9q6d/
>>
>> This is where it all starts, unpacking util-linux:
>>
>> https://postimg.org/image/dd2b549q3/
>>
>> It tried it two consecutive times with same results :(
>>
>> The virtual machine's disk is SATA.
>>
>> Can anybody give me some hint on this?
>>
>> Thanks,
>> Emiliano.
>>
>
>
___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Dng Digest, Vol 23, Issue 42
On Wed, 10 Aug 2016 17:08:20 Simon Hobson wrote: > It does sound like this was an area potentially in want of some looking at. > However, the way to do that would have been to discuss it rationally with the > kernel guys and agree a way forward - putting the fix in well before the > breakage. "Co-operation ? No, never heard of it" seems to be their motto. > Why cooperate when becoming a dictator is s much easier!!! Who cares what havoc you wreak when you have narcissistic personality disorder/psychopathic personality disorder. In other words we'll just put shit out there and let someone else clean up the mess. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
Le 10/08/2016 18:20, Tomasz Torcz a écrit : Current udev_cannot_ be used to populate /dev, it doesn't contain any mknod() calls enymore. Actually mknod isn't necessary since device files are created by the kernel in /sys/block/dev and /sys/char/dev and the hotplugger only needs to copy them to /dev. Didier ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
On Wed, 8/10/16, richard lucassen wrote: Subject: Re: [DNG] vdev - udev is a dead end To: dng@lists.dyne.org Date: Wednesday, August 10, 2016, 11:35 AM On Wed, 10 Aug 2016 16:59:58 +0100 Simon Hobson wrote: > How long before he decides that Grub needs "improving" ? How long before he decides that the kernel needs "improving"? -- richard lucassen http://contact.xaq.nl/ Isn't that what Kay Sievers' kdbus was all about? golinux ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
[DNG] GDBus error in LXDE desktop
Just had this error starting LXDE for the first time: https://postimg.org/image/qr34b416j/ Sorry but it's ok to report bugs on this list? Cheers, Emiliano. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On 08/10/2016 06:11 PM, fsmithred wrote: On 08/11/2016 09:45 AM, aitor_czr wrote: > >Hi fsmithred, > >On 08/10/2016 03:36 PM, fsmithred wrote: >>On 08/11/2016 05:23 AM, aitor_czr wrote: >>> > >>> >Hi Ralph, >>> > >>> >On 08/10/2016 10:46 AM, Ralph Ronnquist wrote: >>Note that I added "loop" to modules because the hardware database is a >>squasfs; I haven't actually verified that it's really needed but just >>left it so. >> >>Ralph. >>> > >>> >All the required modules are included in the kernel: >>> > >>> >SQUASHFS >>> >LOOP >>> >SD_MOD >>> > >>> >Cheers, >>> > >>> > Aitor. >>> > >>No SD_MOD here... >> >>$ grep SD_MOD /boot/config-3.16.0-4-amd64 >>$ >> >> >>-fsr > >SCSI, maybe? > >There are alias... > >Aitor. > > > Yes, lots of SCSI. This one? CONFIG_SCSI_MOD=m -fsr Yes, you will have a: /lib/modules/$KERNEL_VERSION/kernel/drivers/scsi/sd_mod.ko Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Timezones: Was: Your computer clock: Was:Re: vdev
Quoting Simon Hobson (li...@thehobsons.co.uk): > Ah, I stand corrected, according to > https://en.wikipedia.org/wiki/UTC%2B14:00 it was because parts of the > colony were on different sides of the IDL and that meant that they > only had 4 working days in common - Monday on one island would be > Sunday on another, and similarly with Friday/Saturday (in the reverse > direction). I've recently been to one of those places: Fanning Island, part of the Republic of Kiribati (part of the Line Islands). This was part of an ocean cruise from the Port of San Francisco to Honolulu, Fanning Island, Western Samoa, Fija, Vanuatu, New Caledonia, and Sydney -- and one amusement is that my wife's birthday didn't exist on the ship, because we skipped that day in crossing the International Date Line. -- Cheers, Grossman's Law: "In time of crisis, people do not rise to Rick Moen the occasion. They fall to the level of their training." r...@linuxmafia.com http://linuxmafia.com/~rick/lexicon.html#grossman McQ! (4x80) ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On Thu, 11 Aug 2016 11:04:24 +0200 aitor_czr wrote: > > I've now made a snapshot of the vdev files from the working disk. > > available at www.realthing.com.au/files/vdev/vdev-snapshot.tgz. > > Thanks ! Aitor, does your vdev set the permissions and ownerships correctly? -- richard lucassen http://contact.xaq.nl/ ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev - udev is a dead end
- Original Message - From: "Rainer Weikusat" To: Sent: Wednesday, August 10, 2016 12:06 PM Subject: Re: [DNG] vdev - udev is a dead end dev writes: On 08/10/2016 04:26 AM, Didier Kryn wrote: Hello. Thanks to a friendly help, I've found a few mails and articles which deserve to be read: Udev on non-systemd is a dead-end: So.. then.. basically any Linux distro which uses udev to populate /dev/ is going to be S.O.L? Including Slackware presumably? This is seriously old news (dating back to 2012). udev became mandatory on Linux by eating hotplug. systemd is supposed to become mandatory on Linux by eating udev. There are two solutions to this problem: 1) Stop using udev. 2) Fork udev. 3) vdev or eudev | ISMAEL | ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Your computer clock: Was:Re: vdev
At 10/08/2016 11:16 +0200 Svante Signell wote: > Aitor, your computer clock is running ahead again. Can you please sync > with ntpdate regularly in a cron script (or replace the CMOS battery?) Besides having been deprecated for a long time, ntpdate is a bad way to control the clock. You should run NTP instead, which provides much better control of the clock and a nicer load on public NTP servers. Roger ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] vdev
On 11/08/16 00:13, richard lucassen wrote: ... diff -rc clean/usr/lib/vdev/udev-compat.sh echo "event-put $_DEVICE_ID" >> /tmp/udev-compat.log echo "" >> /tmp/udev-compat.log -_RC=$? What are the consequences of moving this _RC? +_RC=$? echo "event-put $_DEVICE_ID" >> /tmp/udev-compat.log echo "" >> /tmp/udev-compat.log Since '$?' is 'the result code of last command", it's IMO best placed immediately after the command whose result code to capture. Here, previously, it captured the result of the (failed, as it were) attempt to echo to the log file, rather than the result of event-put. Ralph. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Bootloaders (was: SystemD's brownie points over non-systemd OSs)
On Wed, 10 Aug 2016 08:01:07 +0100 Simon Hobson wrote: > Peter Olson wrote: > > > I have a machine in that state right now, and rather than try to > > debug it at the Grub prompt, I am just going to reinstall the > > system. > > That's a bit like the old "I'm buying a new car because the ashtray > is full" joke. The closer analogy is "I'm taking a few hours to clean out my whole car because the ashtray is full, and I have a feeling there are ashes under the seats and in the back seat too, and I just want to get on with things." Sometimes a good, prophylactic fresh install is just what's needed. SteveT Steve Litt August 2016 featured book: Manager's Guide to Technical Troubleshooting Brand new, second edition http://www.troubleshooters.com/mgr ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Your computer clock: Was:Re: vdev
On Thu, 11 Aug 2016 11:18:39 +0200 aitor_czr wrote: > Hi Svante, > > On 08/10/2016 11:16 AM, Svante Signell wrote: > > On Thu, 2016-08-11 at 02:00 +0200, aitor_czr wrote: > >> > > >> > Aitor. > >> > > > Aitor, your computer clock is running ahead again. Can you please > > sync with ntpdate regularly in a cron script (or replace the CMOS > > battery?) > > > > Thanks! > > My clock is right: > > aitor@gnuinos:~$ date > Thu Aug 11 11:14:02 CEST 2016 > > It only happens in another computer. My second handed Toshiba works > fine :) > > Cheers, > >Aitor. Hey Aitor, You know how there are people who can't wear a watch because something about them breaks watches? Well, you break the time on computers. I'd recommend both a new cmos battery AND ntp and anything else you can do to keep your computer's time in check, because there's something about you... :-) By the way, I had to fire up my time machine to read this, because you haven't sent it yet. SteveT Steve Litt August 2016 featured book: Manager's Guide to Technical Troubleshooting Brand new, second edition http://www.troubleshooters.com/mgr ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] GDBus error in LXDE desktop
On 08/10/2016 07:50 PM, Emiliano Marini wrote: Sorry but it's ok to report bugs on this list? Cheers, Emiliano. LOL, of course... I think so :) Aitor. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Bootloaders (was: SystemD's brownie points over non-systemd OSs)
Quoting Steve Litt (sl...@troubleshooters.com): > Sometimes a good, prophylactic fresh install is just what's needed. There's something to that. At $FIRM, a big shop where I was Senior Sysadmin for six years in the Operations department, we tried to make every machine auto-buildable using confirmation management (CM) software [1]. If there was even a suspicion of something wonky in a host's software, we would disable it in the hardware load-balancer and re-kickstart it, which installed a minimal OS load, parsed its assigned IP out of the IP/MAC database, IPed the host, and installed/started the CM agent. The latter checked in with the CM master, determined the host's intended role based on its IP, and installed/configured additional software to suit the machine's role. Total downtime for such a rebuild was maybe 1/2 hour. Then, re-enable in the load balancer, and done. If the hardware appeared wonky, same thing except with a swapout for a new host and updating of the IP/MAC records. Fortunately, the presence of the CM agent keeping an eye on things meant _most_ unauthorised changes (e.g., by a coder deciding to go cowboy) would be corrected automatically, but sometimes there's nothing quite like a full rebuild. I really do think making hosts be autobuildable, with all package and conffile state recorded in CM rulesets, is the _right_ way to go for any host that needs to be reliable. I'm aiming to do that in the near future even with machines on my home network. (For a relatively simple CM system suitable for small setups, Ansible is good. https://www.ansible.com/ ) For purposes of my home network, I don't need to make the machines _totally_ automatically buildable, which is a good thing, as I'd rather not deal with d-i pre-seeding, Kickstart, FAI, or that sort of thing if I don't have to (on a small network). Whereas, the gain from CM is _huge_ and worth the trouble IMO. [1] We started out using cfengine 2.x, and like many other shops migrated to puppet. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Bootloaders (was: SystemD's brownie points over non-systemd OSs)
> On August 10, 2016 at 3:01 AM Simon Hobson wrote: > > Peter Olson wrote: > > > I have a machine in that state right now, and rather than try to debug it > > at the Grub prompt, I am just going to reinstall the system. > > That's a bit like the old "I'm buying a new car because the ashtray is full" > joke. Actually it is not. Turns out, I had managed to screw up my partition table. So Grub would never in a million years be able to boot anything off that disk. I found this out during reinstallation when I was presented with a partition table different from what I expected. Nevertheless, my complaint is that Grub is not helpful in these circumstances, lacking even a help command at its prompt. You really need impeccable Grub-fu or another computer with net access. Anyway, the reinstallation worked well, and I will restore the data off my original disk (I was replacing a 20 GB disk with a 160 GB disk when I ran into the problem). Peter Olson ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Bootloaders (was: SystemD's brownie points over non-systemd OSs)
Quoting Peter Olson (pe...@peabo.com): > Turns out, I had managed to screw up my partition table. So Grub > would never in a million years be able to boot anything off that disk. > I found this out during reinstallation when I was presented with a > partition table different from what I expected. It's late for this now, but if you _had_ known that the problem was the partition table, you could have fixed it using TestDisk or gpart http://www.cgsecurity.org/wiki/TestDisk https://en.wikipedia.org/wiki/Gpart (I mention this mostly to spread knowledge of two useful tools.) > Nevertheless, my complaint is that Grub is not helpful in these > circumstances, lacking even a help command at its prompt. You really > need impeccable Grub-fu or another computer with net access. Very true. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Bootloaders (was: SystemD's brownie points over non-systemd OSs)
On Wed, 10 Aug 2016 18:45:18 -0400, Steve wrote in message <20160810184518.2c014...@mydesk.domain.cxm>: > Sometimes a good, prophylactic fresh install is just what's needed. ..aye, todays El Reg: * Bungling Microsoft singlehandedly proves that golden backdoor keys are a terrible idea Updated: Redmond races to revoke Secure Boot policy http://go.reg.cx/tdml/c9288/57d49e7f/ffe47801/2n9H -- ..med vennlig hilsen = with Kind Regards from Arnt Karlsen ...with a number of polar bear hunters in his ancestry... Scenarios always come in sets of three: best case, worst case, and just in case. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Bootloaders (was: SystemD's brownie points over non-systemd OSs)
On Thu, 11 Aug 2016 08:48:47 +0200, Arnt wrote in message <20160811084847.31359...@nb6.lan>: > On Wed, 10 Aug 2016 18:45:18 -0400, Steve wrote in message > <20160810184518.2c014...@mydesk.domain.cxm>: > > > Sometimes a good, prophylactic fresh install is just what's needed. > > ..aye, todays El Reg: > * Bungling Microsoft singlehandedly proves that golden > backdoor keys are a terrible idea > Updated: Redmond races to revoke Secure Boot policy > http://go.reg.cx/tdml/c9288/57d49e7f/ffe47801/2n9H ..and then this: * Linux security backfires: Flaw lets hackers inject malware into downloads, disrupt Tor users, etc Analysis: TCP networking code scores own goal http://go.reg.cx/tdml/c9288/57d49e7f/ffe47801/2nac -- ..med vennlig hilsen = with Kind Regards from Arnt Karlsen ...with a number of polar bear hunters in his ancestry... Scenarios always come in sets of three: best case, worst case, and just in case. ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
