The Linux Questions Thread: a bunch of pitfalls, but technically it's possible

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Linux Questions Thread: a bunch of pitfalls, but technically it's possible

«‹›1027 »

Super Dude: Jan 23, 2005; Do the Jew

Alright something really weird is going on. I stuck in my gparted live cd to format my linux partition to have it ready for Windows 8. However, when it showed the partitions, all it showed was my ntfs partition. I thought something might be wrong with the cd, so I booted into linux, and fdisk on there also only shows the ntfs partition. Where the hell is my linux partition? It obviously exists since I'm sitting here typing this message in Ubuntu.

# ? Sep 14, 2011 03:52

Adbot: ADBOT LOVES YOU

# ? May 2, 2024 10:26

Lysidas: Jul 26, 2002; John Diefenbaker is a madman who thinks he's John Diefenbaker.; Pillbug

Super Dude posted:

Where the hell is my linux partition? It obviously exists since I'm sitting here typing this message in Ubuntu.

Not if you're using Wubi. Your Linux system exists inside a large file on your NTFS partition.

# ? Sep 14, 2011 03:55

Super Dude: Jan 23, 2005; Do the Jew

Lysidas posted:

Not if you're using Wubi. Your Linux system exists inside a large file on your NTFS partition.

Where is it then?

# ? Sep 14, 2011 03:57

angrytech: Jun 26, 2009

hah totally misunderstood the question, nvm

angrytech fucked around with this message at 04:22 on Sep 14, 2011

# ? Sep 14, 2011 04:15

Lysidas: Jul 26, 2002; John Diefenbaker is a madman who thinks he's John Diefenbaker.; Pillbug

I've used Wubi for all of a few minutes, and I'm not inclined to try it in a VM to find out.

https://wiki.ubuntu.com/WubiGuide says that it's in C:\ubuntu. Try uninstalling it from the Control Panel, though.

EDIT: WinDirStat or any similar tool would also tell you this.

# ? Sep 14, 2011 04:18

dolicf: Sep 12, 2010

nitrogen posted:

DNS is overrated, i guess.

To be fair, they're probably advertising the internal resolvers that aren't blocked via DHCP and assuming (and/or not giving no fucks) that most people won't be explicitly specifying their resolvers.

# ? Sep 14, 2011 04:19

NOTinuyasha: Oct 17, 2006; The Great Twist

So I'm dealing with a Debian server which has a mdadm RAID1 with sda/sdb, md0 is boot md1 is swap (deal with it) and md2 is root.

sdb recently popped out for whatever reason so I faulted the disks, removed them, and re-added using all these commands:

code:

sudo mdadm /dev/md0 -f /dev/sdb2
sudo mdadm /dev/md0 -r /dev/sdb2
sudo mdadm /dev/md1 -f /dev/sdb3
sudo mdadm /dev/md1 -r /dev/sdb3
sudo mdadm /dev/md2 -f /dev/sdb4
sudo mdadm /dev/md2 -r /dev/sdb4
mdadm --manage --re-add /dev/md0 /dev/sdb2
mdadm --manage --re-add /dev/md1 /dev/sdb3
mdadm --manage --re-add /dev/md2 /dev/sdb4

This appears to have worked for md0 and md1 but it seems to feel that sdb4 is a 'spare'

code:

/dev/md2:
        Version : 1.2
  Creation Time : Mon Jan 10 10:56:36 2011
     Raid Level : raid1
     Array Size : 1920883498 (1831.90 GiB 1966.98 GB)
  Used Dev Size : 1920883498 (1831.90 GiB 1966.98 GB)
   Raid Devices : 2
  Total Devices : 2
    Persistence : Superblock is persistent

    Update Time : Wed Sep 14 00:27:17 2011
          State : clean, degraded
 Active Devices : 1
Working Devices : 2
 Failed Devices : 0
  Spare Devices : 1

           Name : <removed>:2
           UUID : <removed>
         Events : 10659848

    Number   Major   Minor   RaidDevice State
       0       8        4        0      active sync   /dev/sda4
       1       0        0        1      removed

       1       8       20        -      spare   /dev/sdb4

Where did I go wrong?

# ? Sep 14, 2011 05:44

telcoM: Mar 21, 2009; Fallen Rib

NOTinuyasha posted:

So I'm dealing with a Debian server which has a mdadm RAID1 with sda/sdb, md0 is boot md1 is swap (deal with it) and md2 is root.

sdb recently popped out for whatever reason so I faulted the disks, removed them, and re-added [...]

This appears to have worked for md0 and md1 but it seems to feel that sdb4 is a 'spare'

I guess it's because md1 is still syncing. Perhaps the RAID1 logic is smart enough to know that syncing multiple partitions of the same disk simultaneously would cause silly amounts of overhead (seeking back and forth), or the number of simultaneous on-going sync operations is being restricted.

Every time a disk is hot-added to a RAID set, it begins its life as a spare disk. It should automatically switch to being a regular RAID1 set member as soon as it begins syncing.

# ? Sep 14, 2011 06:37

Thoom: Jan 12, 2004; LUIGI SMASH!

I'm running Ubuntu 10.04 on a headless Gumstix (currently accessing it via CKermit over its USB serial port). It connects to my WiFi network just fine, can download software updates, etc, but I can't connect to it at all over the network. It won't even respond to pings. I'm 99% sure it doesn't have a firewall installed, because I built it with a pretty minimal set of packages. Any idea where to start diagnosing this issue?

# ? Sep 15, 2011 00:27

angrytech: Jun 26, 2009

Thoom posted:

I'm running Ubuntu 10.04 on a headless Gumstix (currently accessing it via CKermit over its USB serial port). It connects to my WiFi network just fine, can download software updates, etc, but I can't connect to it at all over the network. It won't even respond to pings. I'm 99% sure it doesn't have a firewall installed, because I built it with a pretty minimal set of packages. Any idea where to start diagnosing this issue?

How are you trying to connect to it?

# ? Sep 15, 2011 00:49

some kinda jackal: Feb 25, 2003; �
�

ClosedBSD posted:

It should be under Network Connections

On wired click add and then click the IPv4 tab, fill out everything there.

I really wish they would re-label things in there since I frequently click on network tools, which if i recall correctly is like a pretty gui for ping and nslookup or something.

Yeah, I just checked this today. There was definitely no hostname in the ipv4 tab. Oh well. I'm going to flatten this tomorrow and install MacOS anyway, this was just a bit of a test to see how usable I found Ubuntu on a daily basis for a few days.

# ? Sep 15, 2011 00:53

Ninja Rope: Oct 22, 2005; Wee.

angrytech posted:

How are you trying to connect to it?

Also useful:

sudo iptables -L
netstat -an | grep :22

# ? Sep 15, 2011 02:42

Thoom: Jan 12, 2004; LUIGI SMASH!

angrytech posted:

How are you trying to connect to it?

I tried both SSH (ps ax | grep ssh confirms sshd is running) and ping. SSH won't connect, and ping says either "Host is down" or "Route not available" depending on its mood.

iptables isn't installed, but I'll check out what netstat says.

# ? Sep 15, 2011 05:21

angrytech: Jun 26, 2009

Thoom posted:

I tried both SSH (ps ax | grep ssh confirms sshd is running) and ping. SSH won't connect, and ping says either "Host is down" or "Route not available" depending on its mood.

iptables isn't installed, but I'll check out what netstat says.

Check /var/log/syslog as well, specifically from around the time that you try to ping/ssh the system.

# ? Sep 15, 2011 05:48

Craptacular!: Jul 9, 2001; Fuck the DH

Stupid non-technical question: Now that Adobe is dropping AIR support, should I bother looking for alternatives for TweetDeck? Are there any even a fraction as powerful?

# ? Sep 15, 2011 07:41

Ninja Rope: Oct 22, 2005; Wee.

Thoom posted:

I tried both SSH (ps ax | grep ssh confirms sshd is running) and ping. SSH won't connect, and ping says either "Host is down" or "Route not available" depending on its mood.

iptables isn't installed, but I'll check out what netstat says.

Make sure the ip, subnet mask, and default gateway are correct. If they're on the same broadcast domain, host is down means you were unable to resolve the mac address for the given IP address.

# ? Sep 15, 2011 12:13

NOTinuyasha: Oct 17, 2006; The Great Twist

telcoM posted:

I guess it's because md1 is still syncing. Perhaps the RAID1 logic is smart enough to know that syncing multiple partitions of the same disk simultaneously would cause silly amounts of overhead (seeking back and forth), or the number of simultaneous on-going sync operations is being restricted.

Every time a disk is hot-added to a RAID set, it begins its life as a spare disk. It should automatically switch to being a regular RAID1 set member as soon as it begins syncing.

This doesn't appear to be the case, I have re-added the disk and the same thing has occurred twice now. Here's /proc/mdstat

code:

Personalities : [raid1] 
md2 : active raid1 sdb4[1](S) sda4[0]
      1920883498 blocks super 1.2 [2/1] [U_]
      
md1 : active raid1 sdb3[1] sda3[0]
      32127810 blocks super 1.2 [2/2] [UU]
      
md0 : active raid1 sdb2[1] sda2[0]
      499988 blocks super 1.2 [2/2] [UU]

Edit: sdb is failing, that is the issue apparently.

NOTinuyasha fucked around with this message at 20:44 on Sep 15, 2011

# ? Sep 15, 2011 19:17

JHVH-1: Jun 28, 2002

If it tries to rebuild and hits bad sectores on that particular partition it will take it out like that, so the other units show complete.

You can follow something like this howto and swap that sucker out:
http://www.howtoforge.com/replacing_hard_disks_in_a_raid1_array

# ? Sep 15, 2011 22:10

NOTinuyasha: Oct 17, 2006; The Great Twist

Both disks have multiple bad sectors. The rebuild fails because of sda though, which is the mounted drive that it's currently relying on:

code:

[17136768.681782] raid1: sda: unrecoverable I/O read error for block 1491076992
[17136768.727952] md: md2: recovery done.

But yeah, I ran badblocks on sdb and I'm trying another rebuild, it probably wont help though. Unless there's some way to fix sda4 while it's still mounted or something the entire server will probably need to get pulled for major surgery.

# ? Sep 16, 2011 00:39

Longinus00: Dec 29, 2005; Ur-Quan

NOTinuyasha posted:

Both disks have multiple bad sectors. The rebuild fails because of sda though, which is the mounted drive that it's currently relying on:
code:
[17136768.681782] raid1: sda: unrecoverable I/O read error for block 1491076992
[17136768.727952] md: md2: recovery done.
But yeah, I ran badblocks on sdb and I'm trying another rebuild, it probably wont help though. Unless there's some way to fix sda4 while it's still mounted or something the entire server will probably need to get pulled for major surgery.

Having errors on both halves of a raid1 doesn't sound very promising.

# ? Sep 16, 2011 18:02

Bob Morales: Aug 18, 2006; ~~Just wear the fucking mask, Bob~~

I don't care how many people I probably infected with COVID-19 while refusing to wear a mask, my comfort is far more important than the health and safety of everyone around me!

Any good IRC channels? (other than Freenode)

I think I asked this before. SA-specific is nice but I would like one with a larger audience (and n00bs dropping in from time to time)

# ? Sep 16, 2011 19:42

JHVH-1: Jun 28, 2002

I guess you can go hang out with the weirdos on the OFTC network.

Freenode is a network, so how people act in a channel depends on the channel.

# ? Sep 16, 2011 20:38

Bob Morales: Aug 18, 2006; ~~Just wear the fucking mask, Bob~~

I don't care how many people I probably infected with COVID-19 while refusing to wear a mask, my comfort is far more important than the health and safety of everyone around me!

JHVH-1 posted:

I guess you can go hang out with the weirdos on the OFTC network.

Freenode is a network, so how people act in a channel depends on the channel.

Freenode ops/staff flip out on things like swearing or going even a little off-topic.

# ? Sep 16, 2011 20:47

dont skimp on the shrimp: Apr 23, 2008

Bob Morales posted:

Freenode ops/staff flip out on things like swearing or going even a little off-topic.

No, that'd still depend on whatever channel you're in. The IRCops or freenode staff doesn't care about those.

For what it's worth, there are offtopic channels for most distros too, like #archlinux-offtopic on freenode.

E: There's also spotchat which hosts the official linux mint channels, but those don't allow swearing.

# ? Sep 16, 2011 21:25

NOTinuyasha: Oct 17, 2006; The Great Twist

Bob Morales posted:

Any good IRC channels? (other than Freenode)

I think I asked this before. SA-specific is nice but I would like one with a larger audience (and n00bs dropping in from time to time)

irc.synirc.org #yospos

# ? Sep 16, 2011 21:33

Vulture Culture: Jul 14, 2003; I was never enjoying it. I only eat it for the nutrients.

Bob Morales posted:

Freenode ops/staff flip out on things like swearing or going even a little off-topic.

You've basically painted a picture of every Napoleon IRC chanop in history.

# ? Sep 16, 2011 22:16

Nigel Danvers: Oct 29, 2009

Not sure if this would be better suited to the networking thread but I thought I'd try here first.

I've got an Ubuntu box at home acting as the usual jack of all trades server, web/torrent/dhcp/shorewall/etc. It sits between an ADSL modem and the rest of the computers in the house. The ADSL modem is setup as 192.168.2.1 and forwards everything to 192.168.2.2 which is one of servers connections. The servers other side is 192.168.0.2 with all the computers having 192.168.0.[3-9]

Basically this;

pre:

NETWORK          - DEVICE - IP ADDRESS  - SHOREWALL IFACE 
ADSL Modem       - eth0   - 192.168.2.2 - net
Local Network    - eth1   - 192.168.0.2 - loc

This all works great, web servers are externally accessible, ports are blocked, shorewall is masq'ing eth0 to 192.168.0.0 so everything gets internet access.

What I'd like to do next is setup a VPN connection on the server so that all out going internet requests gets routed through that (ppp0) while incoming stuff still hits all the existing servers via the non-VPN ip on the ADSL modem (eth0).

I've managed to get a VPN setup and sort of working. If I create the VPN connect and set it as the default route and then change shorewall to use ppp0 rather than eth0 all internet gets routed through it, but for some reason it now stops responding to incoming connections via internet (eth0).

To get the setup I want, eth0 incoming and ppp0 outgoing should I be using ppp0 as the default route? Or is possible I'm blocking the eth0 incoming in my shorewall somewhere when I set ppp0 as the default?

# ? Sep 16, 2011 22:50

Bob Morales: Aug 18, 2006; ~~Just wear the fucking mask, Bob~~

I don't care how many people I probably infected with COVID-19 while refusing to wear a mask, my comfort is far more important than the health and safety of everyone around me!

Nigel Danvers posted:

pre:

NETWORK          - DEVICE - IP ADDRESS  - SHOREWALL IFACE 
ADSL Modem       - eth0   - 192.168.2.2 - net
Local Network    - eth1   - 192.168.0.2 - loc

Basically ppp0 is a third NIC (think if you added a cable modem to the mix)

You could specify certain traffic to go back through eth0, right now you're routing it all over ppp0. What connections do you want to come through? You could set up rules for a web server or something.

I haven't used Shorewall but I think it would be easy to set up in m0n0wall

# ? Sep 16, 2011 23:09

nitrogen: May 21, 2004; Oh, what's a 217°C difference between friends?

Bob Morales posted:

Any good IRC channels? (other than Freenode)

I think I asked this before. SA-specific is nice but I would like one with a larger audience (and n00bs dropping in from time to time)

irc.synirc.net / #sysadmin
goon/non good ratio is about 50/50.

Just please don't be goony.

# ? Sep 17, 2011 04:05

NOTinuyasha: Oct 17, 2006; The Great Twist

Longinus00 posted:

Having errors on both halves of a raid1 doesn't sound very promising.

The disks are both cheap WD green drives (WD20EARS) so bad sectors might just be normal operation or something. The server is colocated and I really would rather not pull anything unless it bursts into flames. I used hdparm to write to the bad sectors on sda while it was still mounted which seems to have cleared up the problem with rebuilding so I'll work from there. I'll write a cron script to email me if the array gets knocked down to 'degraded' status next time.

# ? Sep 17, 2011 04:08

nitrogen: May 21, 2004; Oh, what's a 217°C difference between friends?

NOTinuyasha posted:

The disks are both cheap WD green drives (WD20EARS) so bad sectors might just be normal operation or something. The server is colocated and I really would rather not pull anything unless it bursts into flames. I used hdparm to write to the bad sectors on sda while it was still mounted which seems to have cleared up the problem with rebuilding so I'll work from there. I'll write a cron script to email me if the array gets knocked down to 'degraded' status next time.

smartmontools is your friend.

# ? Sep 17, 2011 04:17

BlackMK4: Aug 23, 2006; wat.; Megamarm

Jesus. My old server (P4 2.8C, 2x dual port Intel 10/100/1000 NICs, 300gb IDE, 2x 1TB SATA) stopped booting last month after I opened and closed the side panel while it was running after swapping back to my old good PSU (Antec 480w) from a cheap-rear end Raidmax 380w something PSU. (I didn't touch poo poo while the side was off)

Well, I scrounged up parts from my roommate with about the only things carried over being the 2x 1TB SATA drives and PSU. Today I put one of the dual port NICs in the server only to find it will no longer boot and has symptoms of a corrupt boot drive.

What in the poo poo. :suicide:

I can only assume the PSU is hosed up and killing the drives but the rails seemed fine when I watched them.

BlackMK4 fucked around with this message at 07:37 on Sep 17, 2011

# ? Sep 17, 2011 07:28

dolicf: Sep 12, 2010

Where does the boot process stall? Any errors or console output at all?

# ? Sep 17, 2011 09:53

BlackMK4: Aug 23, 2006; wat.; Megamarm

Yes, different errors each time after the GRUB loader relating to file read errors. Sometimes I don't even get to GRUB.

# ? Sep 17, 2011 16:41

Vulture Culture: Jul 14, 2003; I was never enjoying it. I only eat it for the nutrients.

Can anyone point me to a tutorial for Chef that doesn't assume you already know Chef? Everything I'm finding is a lovely little braindump someone did of every other command that they ran.

# ? Sep 17, 2011 17:40

Lysidas: Jul 26, 2002; John Diefenbaker is a madman who thinks he's John Diefenbaker.; Pillbug

I have a ThinkPad X61 Tablet, which has three mouse buttons and a nub between the G, H, and B keys.

Back when I was using Windows, Lenovo had some scrolling emulation software that would kick in whenever you hold down the middle button and move the mouse up and down. This has never worked correctly in Kubuntu, even after installing some synaptics control software and configuring this behavior. It would always revert after a reboot or stop working randomly while the system was running.

After installing a nightly build of Kubuntu 11.10, I tried to scroll on a hunch and found out that it works perfectly without any configuration. Thanks, anonymous KDE and/or Kubuntu developer! :cheers:

# ? Sep 17, 2011 18:36

some kinda jackal: Feb 25, 2003; �
�

I'm looking to put together a small backup solution for a few Linux lab computers due to stupid students overwriting or deleting some report data.

My criteria is that it be able to do incremental backups, and that it have a good UI for restoring individual files, whether that be a Web UI or a client on the end machine. Ideally I'll be making backups to a RAID array of maybe 2TB drives so no tape backup necessary. Also I'd love to be able to restore revisions of files, similar to how I can with Time Machine on my Mac, where I can restore the version of xraydata.csv from a week ago versus from the backup job last night.

I'm looking at Bacula, is there anything else I should be seriously giving attention to?

# ? Sep 17, 2011 18:48

Erasmus Darwin: Mar 6, 2001

Martytoof posted:

I'm looking at Bacula, is there anything else I should be seriously giving attention to?

BackupPC

Bacula's better for a full backup solution, where you're saving stuff to tape, rotating them off-site, and so on. BackupPC works better for an ad hoc backup system that won't necessarily save you from a disaster but which is great for quickly restoring individual files.

# ? Sep 18, 2011 17:31

Thermopyle: Jul 1, 2003; ...the stupid are cocksure while the intelligent are full of doubt. �Bertrand Russell

Take a look at crashplan. I love it. Even if you don't pay for the offsite backup it will backup to other computers.

Crashplan just saved my rear end so I'm feeling pretty good about it...

# ? Sep 18, 2011 17:51

Adbot: ADBOT LOVES YOU

# ? May 2, 2024 10:26

ToxicFrog: Apr 26, 2008

Thermopyle posted:

Take a look at crashplan. I love it. Even if you don't pay for the offsite backup it will backup to other computers.

Crashplan just saved my rear end so I'm feeling pretty good about it...

Crashplan is great provided that you have 1GB+ memory to dedicate to the crashplan background service. I love its features but goddrat it is memory-hungry.

Also it really should have some way of backing stuff up over ssh other than 'mount the remote system using sshfs and back up to that mountpoint'.

# ? Sep 18, 2011 21:13

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Linux Questions Thread: a bunch of pitfalls, but technically it's possible

«‹›1027 »