Archive for the ‘ Linux tips ’ Category

Properly configuring plymouth

Mayhaps this should be better called “getting rid of the blue and white bar loading screen in Fedora 11”, because that was the goal I had in mind when I set out on this mission. I’d like to officially say that [this guy] is just totally incorrect. He reviews Fedora 11 booting as, and I quote here,

“First, the loading screen is a simple blue loading bar. Second, the boot process is extremely quick. Does the first have anything to do with the second? It sure does. The Fedora 11 development team has decided to go with speed instead of looks during the boot process and I think Fedora 11 users will be thrilled with this small change to speed up the LiveCD boot time.”

HE’S CORRECT, I’M SO THRILLED!!! One main reason I find this funny — he assumes he knows exactly what the Fedora development team had in mind when designing this blue bar loading screen, yet he’s not a developer on the team. Nor is he correct. Nor did he even read the release notes or improvements list. Nope, not a single piece of knowledge, data, or documentation to back up this claim.

The truth? The dev team did work very hard on the boot process, and the fruit of their labors is the plymouth codebase. The blue/light-blue/white loading bar? Yeah, that’s actually a fallback when the graphical plymouth booting mechanism fails. Great job doing your homework! So we’ve established two things here, that there’s at least one page on the internet with incorrect information, and that, if you are experiencing the fedora 11 blue bar boot (say THAT one three times fast), you can actually attain a prettier boot process.

This how-to is written both from memory, and specifically for my laptop, which I was attempting to spruce up with a nice graphical boot. If it’s inaccurate, sorry about that, but you can certainly use the base I’m giving you here as a launching point for further research.

First, my laptop has an NVidia GPU in it, so it was necessary to download the proprietary NVidia drivers and install them. I did not install the 32-bit compatibility libraries, since if I do, the NVidia installer says something to the effect of, “Could not check random component #9662, assuming successful installation”. Well NVidia, you know what happens when you assume things… Anyway, once you’ve gotten the nvidia kernel module installed, edit /etc/modprobe.d/blacklist.conf and blacklist the modules “nv” and “nouveau” to prevent them from being loaded. We only want the proprietary driver to be loaded here. Make sure you update /etc/X11/xorg.conf as well and tell it to use the nvidia module.

Next, we need to make the framebuffer have the optimal settings, that way plymouth won’t die and fall back to the blue bar loading screen. Append to the end of the kernel line in /boot/grub/grub.conf the text “vga=ask” and reboot. The kernel will prompt you for a video mode. I recommend typing “scan” here and striking return abruptly, as we want to make sure all available modes are listed. And yes, the abrupt manner with which the enter key is pressed is important. So yeah, pick which mode fits your screen the best, for me it was 1920x1200x32, which had a hex code of like 37D or something. Like I said, this is all from memory. Now, write down this hex code somewhere, because you’ll need it again shortly. After the machine boots, find a calculator and convert the hex code you just entered into the kernel into decimal, then place the decimal value in the /boot/grub/grub.conf file, replacing the “ask” with your number. Mine resolved to decimal “893”, so I ended up with “vga=893” on the kernel line. Congrats, you have now made a hospitable environment with which plymouth can bedazzle you.

Okay, now it’s time to configure the workhorse. We’ll need the appropriate utilities, so run this:

# yum install plymouth plymouth-gdm-hooks plymouth-libs plymouth-plugin-label plymouth-plugin-two-step plymouth-scripts plymouth-system-theme plymouth-theme-charge plymouth-utils

Honestly, most of these were probably already installed, but it never hurts to make absolutely sure. Now, recall before where we compiled the NVidia driver? We’ll need that to be loaded during the boot process so that plymouth can come up. This means that we’ll need a new initrd containing the NVidia module. This is also the purpose of blacklisting the nv and nouveau modules, as they now won’t get included into the initrd. I’ve become Canadian lazy lately, so you can run this command to see how to build a new initrd:

# rpm -aq –scripts kernel | grep initrd

A list of 2 or 4 commands will show up from that. Now find the appropriate one and run it. For me, that was:

# /sbin/new-kernel-pkg –package kernel –mkinitrd –depmod –install

Voila, one new shiny initrd. But the initrd isn’t quite done yet. We need to update plymouth’s hooks into it. First, we’ll want to select a theme for initrd to use. I’ve made sure, at the very least, you’ve installed the “charge” theme in the yum command above. There’s plenty more in the repos, feel free to search for more, install them, and play around. Some are very neat. Anyway, to see which themes are available, do something like this:

# ls -1 /usr/share/plymouth/themes/

As you can see, I have a couple of themes installed, but let’s focus on the charge theme. Set it as the default theme, and then update it’s initrd hooks by running, in order:

# plymouth-set-default-theme charge
# /usr/libexec/plymouth/plymouth-update-initrd

Note: This entire abovesaid process needs to be redone for each NVidia driver or kernel update, otherwise it’ll be broken and fall back to the blue bars again.

Now, reboot and enjoy the pretty loading screen. Now there’s only one issue you’ll need to figure out from here on out that I can’t seem to work out, and that’s when to reboot in order to enjoy the loading screen anyway. I mean, if you followed this tutorial, you’re clearly running Linux, which means you’ll probably not need to reboot any time soon. Effectively, you may have just wasted a good 15 minutes setting up a boot screen you’ll never really enjoy too much. Thanks for reading my blog, though, I do appreciate it. I promise next time I’ll include a spoiler warning at the top of the article. Have a nice day! 🙂

64-bit Adobe Flash Player

It exists! It’s not in the Adobe yum repos yet, as it’s still an alpha build, but you can download it here:

[Native Linux 64-bit Flash Player from Adobe Labs]

It works very well. No longer do you need to get nspluginwrapper working, and then configure pulseaudio (if that’s your distro) to listen to 32-bit apps, blah blah blah. Simply unpackage, throw the .so file into the firefox plugins directory, and restart firefox. Great stuff. I’ve also noticed there’s a performance gain to be seen, my flash videos used to lag and flicker on youtube when I made them fullscreen, and this doesn’t happen anymore. Sound worked out of the box with pulse.

And the best part, although it’s supposedly an alpha build, it crashes a heck of a lot less than it used to. No more npviewer.bin segfaults in the /var/log/messages file! My guess is that this is because of the lack of a hacked wrapper. But, that being said, if it crashes on you a lot after trying it, don’t be too upset. At least they’re trying.

I’m the fool. I took the plunge this week and installed a fresh clean copy of Fedora 11, but I made my root partition btrfs instead of something more… well, for lack of better words, stable and mature. This should serve as a little documentation of my findings, in the hopes that others don’t make the hasty call to dedicate their root partition to the new filesystem.

From the dawn of my days with Linux, I have been told to heed warnings that this new system gives. This isn’t Windows any more, the warnings that you receive are no longer benign. My first experience with this was when I had a hard drive that was dying, so I decided to do a fsck. The only problem was that I did a fsck on the root partition – while it was still mounted — read/write. In all fairness, it was my own arrogance that made me add the flag to manually override that protection system. I figured, “what could get worse? The drive is already dying”. I did this with no backups. Three days and 20-something manual superblock hex-edits later, I finally got my data back. The valuable thing from that incident is the experience, though. I now knew how to do manual superblock recreations, which always comes in super handy. I also had learned to heed the warnings of the core developers very well.

Until Saturday. My friend had told me before about how funny the new anaconda argument is: “icantbelieveitsnotbtr”. And now who says these hardcore developers don’t have a sense of humor? Well, I’ve learned a heck of a lot over these last few years, and sometimes I even feel guilty that I don’t give back as much as I should to the open source community. So I figured I’d give their hard work a shot. I’ve heard nothing but glowing reviews about ZFS, and btrfs is supposed to be the up-and-coming big thing in the filesystem world. I figured, at the very least, I could do some benchmarks or something.

So here goes nothing…

I fired up the installer, adding the appropriate flags. I had /dev/sda1 already formatted 50 gig ntfs for Windows. This was to be a dual-booting system. I added a 512-meg /boot, sda2, and a 2-gig swap, sda3. sda4 was to be a pointer to an extended partition table, where sda5 resides, filling up all of the remaining space of 59 gigs. I picked my packages and started the installer. Uh oh, the format of sda5 as btrfs failed. Oh well. Reboot, repeat. Same results, but this time it just died with some non-helpful error code 1. Uhhh. So I tried to boot again. My partitioning table was simply gone. WTF.

I remade the partition table, and figured, “third time’s the charm!”. I selected btrfs as the filesystem for /. If this didn’t work, I was going to give up and use the (now-default) ext4 filesystem. Alas, it worked. I was actually looking forward to using the new filesystem. The install finished normally, and I booted into my new Linux system.

First thing’s first, I ran a full yum upgrade. I noticed something just then. The seek times with btrfs were amazing, read speeds were okay, but the write speeds were flat-out terrible. I’m guessing that’s simply because of all of the metadata btrfs stores, but I honestly have no idea why that is. Oh well, I do far more reads than writes, so this works out well. I then proceeded to use the system for a while. After my used disk space started to get above 25%, I started to get really unexpected results – kernel failures, corrupted filesystem blocks, half-missing extent entries, system bus errors, the whole nine yards.

I did the next logical thing, I strace’d the processes that were dying unexpectedly on me. Worst of all was yum, it would just die with a “Bus Error”. Strace revealed it was hitting massive ENOSPC (no space left) errors trying to open files on the hard drive. df reported the filesystem as exactly 27% full. A google search revealed this as a known bug, but no one else has really had the problem until their filesystem reached over 75% full. So why was mine being the opposite? I theorized that it may have been because of loads and loads of useless metadata across the hard drive. I rebooted (the only thing that fixed the ENOSPC errors) and installed the btrfs-progs package. This package includes userspace utilities needed to create, check, modify and correct any inconsistencies in the btrfs filesystem. I ran a btrfsck, which reported no problems. I read that it may be useful to re-examine my metadata by forcing a balance of it across all drives, even though it only existed on one partition. This is done with the command “btrfs-vol -b /”. Kicker – this operation, after about 5 minutes of heavily accessing the drive, segfaulted. I was left with a half-balanced btrfs system, ENOSPC errors everywhere, and a feeling like this was the beginning of the end.

I didn’t quite want to give up yet, but I also wanted a usable machine. I had read somewhere that kernel 2.6.30 had an insane number of btrfs improvements. On a personal wish, I downloaded the kernel source for 2.6.31 and compiled. Booting into the new kernel was like a breath of fresh air.

SIDE NOTE: kernel > 2.6.30 is awesome. readahead is now merged into the kernel source for MUCH faster booting. There have been a ludicrous amount of improvements, check the changelog on for a good list of these changes.

Anyway, the new kernel loaded up just fine. The errors held off for a bit, long enough to make me think they were gone. But then they came back full-force. The balance segfaulted again. The fsck said nothing was wrong. No one had any idea what was going on. I do like to help the open source community, but I simply don’t have time at the present to dig through hundreds of thousands of C code source lines in order to fix a bug this serious. After filing 4 bug reports, submitting probably close to 50 kernel errors, and genuinely giving it my best college try, I was done with btrfs. I rebooted into anaconda and formatted my system as ext4 and went along my way. Yes, I know you can *theoretically* convert a btrfs volume to ext4, but with the structure so far destroyed, I didn’t even bother wasting my time.

Moral? I guess just know what you’re getting into if you decide to give btrfs a try. When the kernel developers say that a feature is experimental, take that literally. There is no warning here, only some sound advice – proceed at your own risk, make plenty of backups, and just generally don’t expect too much of the filesystem at this early stage of it’s development.

Okay so as part of being a Linux sysadmin, I’ve run into the need to rebuild a software RAID (mdadm) array. This process usually takes forever, and in an effort to make sure it won’t take any longer than necessary, I usually leave the volume in question unmounted during the resync/reshape. I know, I know, you can read/write to an array during the process safely, otherwise anaconda (the most superior Linux installer) wouldn’t be able to move forward with the install for a long while. But I still like the reshape to go faster. I’d rather not use the system for an hour and have it be perfectly fast afterwards than have the machine crawl along all day.

Long-winded ideals speech beside, I want to monitor how long the rebuild is going to take in quasi-realtime. Details of the software raid can be found in the file /proc/mdstat. You can cat this file once in a while to see the progress. But that wasn’t ever good enough, too much of that whole keyboard-button-pushing thing in order to get the statistics to update. So I would usually type this in real quick:

while [ 1 ]; do clear; cat /proc/mdstat ; sleep 1; done

It’s just a simple loop that keeps the information fresh on the screen. Well boy did I just find out something today that helps greatly with that dilemma. Enter the Linux “watch” command, courtesy of the procps package. The watch command automatically updates the information on the screen in the command that follows the binary on the shell. For example, the command:

watch cat /proc/mdstat

will update the file exactly like the loop does. But watch does it even better, as it prints a clock in the corner and lets you know how often it is updating. Plus, watch doesn’t make the screen flicker for a millisecond like the shell clear does. The default delay in updating is 2.0 seconds, but you can change that easily via the -n flag. For example, to update every second, as in the custom loop above, try the command:

watch -n 1 cat /proc/mdstat

and be amazed. It would seem at first glance that the delay can be set as low as 0.1 seconds. Pretty neat. It’s an interesting tool, and certainly one that is very useful for these sorts of things. Check out the man page for more information on the command.


Quick tip. You’re working on setting up a server late at night. It’s dark out. Your roommate is sleeping. And every time you push the damn tab button, the computer makes a beeping sound. WHAT THE HELL?!

Silence the PC Speaker with “rmmod pcspkr”. Make it loud again with “modprobe pcspkr”.

Your roommate will thank me for this… whether he/she knows it or not 😀