|
Sounds good! I'm mostly RHEL, I noticed upstart in RHEL 6 and was able to avoid it, but now that 7's gonna have systemd I figured I was time to update my skills.
|
# ? Jul 6, 2013 16:49 |
|
|
# ? Jun 2, 2024 19:08 |
|
hackedaccount posted:Sounds good! I'm mostly RHEL, I noticed upstart in RHEL 6 and was able to avoid it, but now that 7's gonna have systemd I figured I was time to update my skills. This leads into my usual admonition: if you want to keep your skills up to date as RHEL admin, you should be using Fedora at home, and potentially as a workstation. There is no better way to see the direction RHEL is going in, the skills you should have, and the caveats you should know about.
|
# ? Jul 6, 2013 16:52 |
|
hackedaccount posted:Hey it looks like systemd has been confirmed for RHEL 7: https://access.redhat.com/site/solutions/417213 Upstart is dead. As for systemd: if we expect RHEL 7.0 to be half the disaster that RHEL 6.0 was, you might be waiting another couple years before seeing it in the wild... and by then systemd might not be mandatory.
|
# ? Jul 6, 2013 17:30 |
|
Notorious b.s.d. posted:As for systemd: if we expect RHEL 7.0 to be half the disaster that RHEL 6.0 was, you might be waiting another couple years before seeing it in the wild... and by then systemd might not be mandatory. .0 releases are almost always a "disaster". .1 is fine. This means you'll be waiting ~6 months before likely adoption of RHEL7, as an average business using RHEL. And you'll probably be using 7.0 on a few non-production systems for VMs for testing your use cases. Moreover, systemd is not going away. It's not mandatory (systemd can use SysV scripts without modification), but there's little indication that there'll be another init system replacing systemd anytime soon.
|
# ? Jul 6, 2013 17:35 |
|
For whatever reason I keep installing the old edition of Ubuntu 12.04 with the 3.2 kernel and I'd love to move up to 12.04.2 with 3.5 kernel. I've hit multiple issues already with 3.2 with my ethernet card not being recognized and my wifi dongle not working with USB 3.0 ports, so it'd be a good idea to upgrade. Now, how do I do this without breaking everything? Is there a safe upgrade route? Anything I can do to rollback in case of massive fail?
|
# ? Jul 7, 2013 05:02 |
|
I need a way to do full screen Mono applications on 12.04. The application is a borderless C# application that goes full screen on loading. It works great, except that full screen doesn't actually mean full screen on Ubuntu it seems. The launcher and menu bar are always presentwith Unity. I've set the launcher to auto hide, and the app takes up that space just fine, but there seems to be no way to get rid of the menu bar or get the app on top of it. Is there something I could do to get my Mono app to be truly full screen? Would switching to Gnome get me to my goal? Would I run into the same problem if this weren't a Mono C# app and if this were in, say, Python or Java?
|
# ? Jul 7, 2013 05:46 |
|
DreadCthulhu posted:For whatever reason I keep installing the old edition of Ubuntu 12.04 with the 3.2 kernel and I'd love to move up to 12.04.2 with 3.5 kernel. I've hit multiple issues already with 3.2 with my ethernet card not being recognized and my wifi dongle not working with USB 3.0 ports, so it'd be a good idea to upgrade. There is zero danger in compiling your own kernel, which is probably the route I'd take. Does 3.5 come with 12.04.2? If not, grab 3.9 (or whatever) sources, zcat /process/config.gz >/usr/SRC/linux/.config to get whatever the Ubuntu defaults are. Some options have probably changed, but not many. Mortanis posted:I need a way to do full screen Mono applications on 12.04. The application is a borderless C# application that goes full screen on loading. It works great, except that full screen doesn't actually mean full screen on Ubuntu it seems. The launcher and menu bar are always presentwith Unity. I've set the launcher to auto hide, and the app takes up that space just fine, but there seems to be no way to get rid of the menu bar or get the app on top of it.
|
# ? Jul 7, 2013 14:23 |
|
DreadCthulhu posted:For whatever reason I keep installing the old edition of Ubuntu 12.04 with the 3.2 kernel and I'd love to move up to 12.04.2 with 3.5 kernel. I've hit multiple issues already with 3.2 with my ethernet card not being recognized and my wifi dongle not working with USB 3.0 ports, so it'd be a good idea to upgrade. Install the linux-generic-lts-quantal package for 3.5, or linux-generic-lts-raring for 3.8.
|
# ? Jul 7, 2013 15:15 |
|
Lysidas posted:Install the linux-generic-lts-quantal package for 3.5, or linux-generic-lts-raring for 3.8. It's amazing that Ubuntu does this. Why would you potentially break abi compatibility on LTS?
|
# ? Jul 7, 2013 15:31 |
|
Is there a way to preserve console text coloring when using "less" command? e: I mean when piping to less, such as "ls -l | less" peepsalot fucked around with this message at 23:49 on Jul 7, 2013 |
# ? Jul 7, 2013 21:57 |
|
peepsalot posted:Is there a way to preserve console text coloring when using "less" command? Looks like this is a 2 part problem because of the two sides of the pipe. First part is that ls doesn't pass color information through a pipe if its set to auto. Second would be to get less to process it. So here you go: ls -l --color=always | less -R
|
# ? Jul 8, 2013 00:31 |
|
Lysidas posted:Install the linux-generic-lts-quantal package for 3.5, or linux-generic-lts-raring for 3.8. What does that mean for the existing packages on the system? Is it assumed that most will work without need for changes and some will get an upgrade if apt detects one?
|
# ? Jul 8, 2013 00:31 |
|
DreadCthulhu posted:What does that mean for the existing packages on the system? Is it assumed that most will work without need for changes and some will get an upgrade if apt detects one? If one of your packages requires a kernel driver (Oracle RAC, Clearcase, VirtualBox whatever) or links directly against libraries which come with X.org (because they apparently update that as well) that's not part of the mainline kernel, it will probably break, which seems really intolerable for an LTS distro, but :ubuntu: This is why people use RHEL/CentOS/whatever in production -- kernel ABI for 10 years.
|
# ? Jul 8, 2013 02:26 |
|
evol262 posted:If one of your packages requires a kernel driver (Oracle RAC, Clearcase, VirtualBox whatever) or links directly against libraries which come with X.org (because they apparently update that as well) that's not part of the mainline kernel, it will probably break, which seems really intolerable for an LTS distro, but :ubuntu: This is why people use RHEL/CentOS/whatever in production -- kernel ABI for 10 years. Packages of that sort within Ubuntu usually rely on DKMS to rebuild modules. DreadCthulhu posted:What does that mean for the existing packages on the system? Is it assumed that most will work without need for changes and some will get an upgrade if apt detects one? If it's something from a third-party repo you might be out of luck.
|
# ? Jul 8, 2013 03:24 |
|
hifi posted:Packages of that sort within Ubuntu usually rely on DKMS to rebuild modules. The point was that DKMS won't be able to rebuild those modules if the ABI changes, e.g. the version of virtualbox supplied by default in 12.04 won't build on 3.8. Of course those kernels are completely optional so I don't see what the issue is. If Canonical was forcing those kernels on every LTS system that would be another thing entirely.
|
# ? Jul 8, 2013 04:03 |
|
Stupid question, and sorry if it's been covered before: My mac friends have made me jealous with their "open" command-line utility which will find the proper program to open whatever file you specify. For example: "open mypaper.pdf" would launch Adobe Acrobat and load the file "my.pdf", and "open myporn.avi" launches whatever plays porn on a mac, etc. Does such a thing exist for Linux? Specifically I am using Mint-DE, so a package from them would be great, any old .deb would be fine, or source would be A-ok too.
|
# ? Jul 8, 2013 12:41 |
|
Try xdg-open?
|
# ? Jul 8, 2013 13:01 |
|
emf posted:Stupid question, and sorry if it's been covered before: I use KDE on my Gentoo PC. If you have nepomuk running (might be called semantic desktop), you can hit Alt+F2 and type the name of your file into KRunner. Because xdg-open/gnome-open/etc exist, I would think there's a GNOME equivalent of this, but I can't say for sure. Or you can just use xdg-open from the command line.
|
# ? Jul 8, 2013 13:06 |
|
Longinus00 posted:The point was that DKMS won't be able to rebuild those modules if the ABI changes, e.g. the version of virtualbox supplied by default in 12.04 won't build on 3.8. Of course those kernels are completely optional so I don't see what the issue is. If Canonical was forcing those kernels on every LTS system that would be another thing entirely. Not really as optional as presented: Ubuntu Team posted:In an effort to support a wider variety of hardware on an existing LTS release, the 12.04.2 and newer point releases will ship with an updated kernel and X stack by default. These newer hardware enablement stacks will be comprised of the newer kernel and X stacks from Quantal, Raring, and S (name TBD). These enablement stacks are only intended for use on x86 hardware at this time. Those running virtual or cloud images should not need these newer enablement stacks and are thus recommended to remain on the original Precise stack. To remain on the original Precise stack there are a few options: Putting an ABI-breaking update into a minor point release is wholly unacceptable for businesses. I know we're not perfect about backporting hardware enablement either (though in general, support for new server-class hardware is available until the end of production 2), but what do you think Oracle or IBM would say if you sent them a 12.04.2 ISO that installs 3.5 by default for recertification?
|
# ? Jul 8, 2013 15:30 |
|
I'm having trouble getting menu colors in Gnome 2.x to be readable No matter what settings are chosen, this is what the Gnome menu and some apps like Firefox turn into. A user is getting kind of mad about it so any idea why these menus would be different from all others? Yes that's a photo of a screen. Prntscrn doesn't work when a menu is open.
|
# ? Jul 8, 2013 16:05 |
|
Dead Inside Darwin posted:I'm having trouble getting menu colors in Gnome 2.x to be readable It looks like a problem with gtkrc. You may want to check ~user/.gtkrc* and /etc/gtk/gtkrc on his system against other users. GNOME/GTK issues aren't my strong point, though.
|
# ? Jul 8, 2013 16:29 |
|
Dead Inside Darwin posted:I'm having trouble getting menu colors in Gnome 2.x to be readable In the appearance panel you should be able to check other themes, and make sure the user has no ~/.gtkrc or anything like that. Back it up if he intentionally modified it so he doesn't complain when all his colors are wrong, of course.
|
# ? Jul 8, 2013 18:27 |
|
Suspicious Dish posted:In the appearance panel you should be able to check other themes, and make sure the user has no ~/.gtkrc or anything like that. Back it up if he intentionally modified it so he doesn't complain when all his colors are wrong, of course. All .gtkrc files are basically empty I've had him try to change things in the apperance panel and he claims it doesn't change anything. He's saying its specifically the "Red Hat" menu, Firefox, and xterm.
|
# ? Jul 8, 2013 19:12 |
|
Does he use any weird focus settings? It could be an issue where the panel or Firefox is somehow getting applied the "unfocused" theme by accident. Try using Ctrl+Alt+Tab to go to the panel and see if it still happens. xterm does not use GTK+, so I'm going to have to see a screenshot of that one.
|
# ? Jul 8, 2013 19:19 |
|
JHVH-1 posted:Looks like this is a 2 part problem because of the two sides of the pipe. First part is that ls doesn't pass color information through a pipe if its set to auto. Second would be to get less to process it. Neat, thanks. By the way how does ls know if its being piped?
|
# ? Jul 8, 2013 20:28 |
|
isatty, a function in the standard library.
|
# ? Jul 8, 2013 20:33 |
|
Is there a way to tell postfix to not log or spit out the recipient in any logs it generates? Or at the very least have it obscure everything before the @, leaving the domain is fine. An obscure IA requirement just slapped me around
|
# ? Jul 8, 2013 20:56 |
|
Suspicious Dish posted:Does he use any weird focus settings? It could be an issue where the panel or Firefox is somehow getting applied the "unfocused" theme by accident. Try using Ctrl+Alt+Tab to go to the panel and see if it still happens. There's no focus settings in play There was a line in one of the gtkrc's pointing to a specific scheme's gtkrc, I commented it out and now Firefox and other GTK apps look fine, but terminal and the Gnome menu still look bad.
|
# ? Jul 8, 2013 21:01 |
|
evol262 posted:Kernel crashdump/kdump. Thanks, but nothing made a difference, and nothing I tried helped me get anymore info. I tried pulling the gfx card just in case and it was still happening. I switched in an Intel DC S3500 ssd in case it was the Crucial M4 causing this. I kinda hope it was as I broke the Crucial removing it. So far no crashes. e: New and different failure mode: Jul 8 20:19:04 james-lenovo kernel: [14561.339303] ehci-pci 0000:00:1d.0: fatal error Jul 8 20:19:04 james-lenovo kernel: [14561.339309] ehci-pci 0000:00:1d.0: HC died; cleaning up Jul 8 20:19:04 james-lenovo kernel: [14561.339345] usb 2-1: USB disconnect, device number 2 Jul 8 20:19:04 james-lenovo kernel: [14561.339348] usb 2-1.3: USB disconnect, device number 3 Jul 8 20:19:04 james-lenovo kernel: [14561.346551] usb 2-1.4: USB disconnect, device number 4 I think this system is just cursed. Aquila fucked around with this message at 04:21 on Jul 9, 2013 |
# ? Jul 9, 2013 02:11 |
|
I usually get that error when trying to use a failing/failed USB drive. Do you have any such devices connected to your PC? (Possibly even a printer with a built in card reader)
|
# ? Jul 9, 2013 04:26 |
|
I've got a question on checking file access times. I'd just like to see if an application is reading from files by if their atimes update, but the system I'm running on is mounted as a relatime system, and I can't change that. In my understanding, that means that the access time updates if the file is read AND one of the following:
I figured I'd just manually set the access times on files to the distant past, so I could check them post application-run and see what's been read. This works if I find some random file that's been sitting on the OS for a while and stat it to see the access time and then cat it: code:
code:
Edit: I just realized that the stat actually changed between those last two by a few minutes, though it's still set a month in the past ... I'm guessing I'm really messing up the underlying data somehow. I can work with that, though - I just needed some way to check it's changing. Still curious if anyone has any thoughts on it. Junji Eat More fucked around with this message at 07:05 on Jul 9, 2013 |
# ? Jul 9, 2013 06:56 |
|
Necronomiconomist posted:I've got a question on checking file access times. I'd just like to see if an application is reading from files by if their atimes update, but the system I'm running on is mounted as a relatime system, and I can't change that. In my understanding, that means that the access time updates if the file is read AND one of the following: That aside, inotify is a much better solution for your problem.
|
# ? Jul 9, 2013 17:10 |
|
Seth Vidal (skvidal on IRC), the creator of yum and a dear friend at Red Hat, was killed in a hit-and-run accident today. Rest in peace, good buddy.
|
# ? Jul 9, 2013 20:42 |
|
Any advice for someone installing Mint for the first time on his desktop during his once every other year "Let's check out how Linux on the desktop is coming along" spontaneous urge?
|
# ? Jul 9, 2013 22:04 |
|
El Generico posted:Any advice for someone installing Mint for the first time on his desktop during his once every other year "Let's check out how Linux on the desktop is coming along" spontaneous urge? Virtualize it.
|
# ? Jul 9, 2013 22:33 |
|
How do I prevent linux from automatically mounting disks/partitions it detects filesystems on? For example, I need to partprobe /dev/vda to make the partition nodes show up in /dev/ but absolutely do not want to mount the fs detected on partition 4 in /media/vda4 I'm only running busybox in 3.9.5
|
# ? Jul 10, 2013 16:11 |
|
Illusive gently caress Man posted:How do I prevent linux from automatically mounting disks/partitions it detects filesystems on? For example, I need to partprobe /dev/vda to make the partition nodes show up in /dev/ but absolutely do not want to mount the fs detected on partition 4 in /media/vda4 Linux doesn't do this. Gnome does this. KDE does this. Some rescue disks do this. It won't get mounted by default. If this is happening, please provide more details so you can get an actual answer.
|
# ? Jul 10, 2013 16:20 |
|
evol262 posted:Linux doesn't do this. Gnome does this. KDE does this. Some rescue disks do this. It won't get mounted by default. If this is happening, please provide more details so you can get an actual answer. It's literally just busybox running on an ELDK minimal ramdisk with like SSH and poo poo added. If there's something I've put on there that would cause this, I have no idea what it would be.
|
# ? Jul 10, 2013 16:47 |
|
Quick and stupid one, I'm trying to grep a log file for "201" but getting all lines matching due to "2013" in timestamps. Any way to not match 2013 while still allowing any other occurrence of 201 in the same line to match? Somehow I get the feeling there's a regex in this answer, and that's about where my brain goes on vacation.
|
# ? Jul 10, 2013 16:58 |
|
|
# ? Jun 2, 2024 19:08 |
|
wolrah posted:Quick and stupid one, I'm trying to grep a log file for "201" but getting all lines matching due to "2013" in timestamps. Any way to not match 2013 while still allowing any other occurrence of 201 in the same line to match? Edit stupid answer out -- without knowing exactly what you want to still match this would screen out just the lines with the timstamp and no other 201: code:
covener fucked around with this message at 17:17 on Jul 10, 2013 |
# ? Jul 10, 2013 17:03 |