Stu's questions

I found instructions on how to create a virtual disk for use with virtualbox that points to a raw block device. You can't do it from the ui, but you can do it from the command line:

VBoxManage internalcommands createrawvmdk -filename "</path/to/file>.vmdk" -rawdisk /dev/sda

this works great except it doesn't do direct i/o to the block device so everything ends up in the page cache, and if you know how the page cache works, it spoon feeds the block device dirty pages one at a time. This takes forever on a slow block device.

If virtualbox accessed the block device with direct io then batched reads and writes would go directly to the block device as a batch (a bio with many segments in it) and it would all go a lot more quickly.

I can't find an option anywhere that describes how to do this or if it is even possible. Anybody know of such a thing?

Basically I want to do something like strace ssh user@host and instead of seeing the system calls, I want to watch the tcp connection packets as the connection is established.

I'm having this weird problem between two specific machines, where b can connect to a but a can't connect to b most of the time, and when it is able to connect the connection is very slow and it takes a minute just to get the ssh greeting.

Sounds like packets being dropped, so I'm looking for a way to diagnose. I know I can fire up wireshark or something (suggestions welcome) but I figured if there was something simple and specific to a process that would be handy.

I'm new to docker and I'm playing with the network options. bridge networks makes a private network and bridges it to the host's network.

The host network uses the host machine's network stack, your docker processes live in the same network space as processes on the host machine.

What I'm looking for is something like what virtualbox does when it makes a what it calls a bridged network adapter. It is a peer of the host's network but the interface is separate and can get its own ip address from the network's dhcp server.

Does functionality like that exist in docker? I can't find it, if it does.

Thanks.

Exact opposite of this question:

How to makesure VM is in Para virtualization mode on XEN?

I've got xenserver 6.2 and nowhere in the UI when creating a VM does it give me the option to use HVM and not paravirtualize by default.

Any ideas?

I just moved all my web applications to a really small vm (128 megs of memory) running linux xubuntu, with no GUI. Most of my applications run fine (if a bit slow, but that's acceptable) except for one. That one is a very small lightweight cgi application. Wordpress (php) and another jsp application run fine, but when I run the cgi, most of the time I get service temporarily unavailable. But sometimes, it works fine.

The vm is not swapping, and it's otherwise responding quickly. cpu is 90%+ idle. load average is nothing... 13:26:31 up 9 min, 2 users, load average: 0.00, 0.33, 0.34

How can I find out what's making apache decide to kick back a 503? There's nothing in any log. It doesn't even show an attempt to hit the url.

I have no idea where else to look. Any suggestions would be appreciated.

I'm looking for a command line tool that can install a package from a file and automatically resolve dependencies via download. Does such a thing exist?

I use postfix to do final delivery for a mail system I've written. So this postfix installation only accepts mail from me, and bounces. When the postfix queue is empty, it can accept a piece of mail in 5ms or so. When the queue has 150k-200k messages or so, the handoff speed from me to postfix is really slow, like 50-100 sometimes 500ms.

I have log times on every piece of the SMTP conversation, so I can see where the time is going. Some of it is when I'm waiting for a response from the RCPT TO command, but the vast majority of the time disappears after I send the final dot, before I get the "250 okay queued as..." response.

I'm thinking maybe the RCPT TO delay is in a DNS lookup, but that doesn't help me with the queuing time. I keep one TCP socket open for all the email I'm sending, and I RSET the conversation for each mail, so there's no time lost to making and breaking TCP connections, it's all waiting for postfix to queue the message.

Everything I've read about performance tuning postfix has to do with queue management and sending mail for this and that domain here and there. But I'm concerned with getting postfix just to accept the mail quickly in the first place.

Is there any way to find out what postfix is doing with all this time or is there a way to make it go faster? The /var/spool/postfix queue is on local disk, so I can't make that any faster.

Any suggestions?

Solaris would be preferable, but I'll take linux or something I can compile. Does such a beast exist?

We're using virtual_alias_domains to tell postfix we handle the domain and virtual_alias_maps to rewrite the address we want to forward to.

But in some cases, the rewritten address has the same domain, but we don't handle it (and we're not the mx for that domain, the mail gets forwarded to us in the first place via a hardcoded method that doesn't do mx lookups).

So basically I want postfix to accept the mail for the domain example.com do a virtual_alias_maps lookup, which returns (for example) [email protected] and then forward the mail to the real mx for example.com.

I've tried a bunch of things, but it seems as long as example.com is in our virtual_alias_domains list (which it has to be or postfix won't accept the mail in the first place) I can't get it to try and send the email out via the real mx.

Any suggestions?

I used to have a RAID 0 setup but I ditched it and restarted from scratch with a plain non raid single disk system. The second disk is just used for general storage.

Well today, I boot the machine and for whatever reason the BIOS booted from the secondary drive which still had the old RAID 0 boot record and OS and everything. Took me a few minutes to figure out what went wrong, I changed the boot order in my BIOS and my 'new' machine came back.

How do I blow away the bootability of the old secondary drive without destroying any of the partitions on it? The MBR has the partition table so I can't just blow it away, but I can't find any partition tool that will erase the MBR while keeping the partition table.

This is a ubuntu 10.10 setup.

I want connections coming in on ppp0 on port 8001 to be routed to 192.168.1.200 on eth0 on port 8080.

I've got these two rules

-A PREROUTING  -p tcp -m tcp --dport 8001 -j DNAT --to-destination 192.168.1.200:8080

-A FORWARD -m state -p tcp -d 192.168.1.200 --dport 8080 --state NEW,ESTABLISHED,RELATED -j ACCEPT

and it doesn't work. What am I missing?

I have a windows xp profession sp2 machine which is my corporate desktop, I do all my development on it. I have to reboot it every 3-4 days because slowly crss.exe starts eating more and more cpu, and as it must have something to do with the GUI, the whole interface slows down and becomes unusable. I've searched the web to no avail. It doesn't use 100% cpu, it becomes unusable long before that.

Anybody ever see this one?

The IT guys here gave it a looksee but they had no idea so they gave up.

I have a system with symlinks EVERYWHERE, so given a particular directory, is there a simple way to find out what mountpoint this directory is on? Particularly interested in solaris.

I can mount an nfs share on my linux machine just fine, but this is a macbook pro that is doing the mounting, and when the laptop goes away from my home network nfs hangs trying to communicate with the mount when I try and access it.

What I'm looking for is a way to mount it automatically when I'm on my network and disconnect when it goes away. Is this possible?

I see that mysql replication can synchronize from sybase, but can sybase replication server write to and read from sybase? I'm trying to figure out a migration path of a big old legacy sybase system lots of little applications to mysql and figure I can move the newer apps sooner than the older ones so I want to be able to use both and have sybase replication do the synchronizing. Is this possible?

It sounds simple enough, but I'm having a hell of a time getting it to work. Apache expects the first line of my script to say #!<path_to_cmd_shell> And I don't want to use perl or anything else, I want to use c:\windows\system32\cmd.exe But no go.

When I run the rest of the cmd file from the cmd prompt, it works as expected.

I even have a test cgi that does this:

@echo off
echo Content-Type: text/html
echo.
echo ^<html^>^<head^>^</head^>^<body^>hi from stu the cgi ^</body^>^</html^>

and that works.

But when I change the script to run another exe (whose output would be the cgi response) I get all sorts of premature end of script headers.

I'm guessing that it has something to do with the output of the exe not piping back through the shell's output.

Any idea how to make this go?

I know I can use samba or nfs, but which is better for two linux machines to talk to each other or is there a better one than those two?

I'm trying to make a raid array on an existing linux ubuntu install.

I'm following this tutorial... http://howtoforge.org/software-raid1-grub-boot-fedora-8

After going through the list of things a million times I finally understand what's going on. You make the raid device, on your new blank drive, copy your old / drive to it, set up the grub menu.lst, fstab, mtab initrd and grub MBR to all point to the raid device (which I have defined and is working) and then you reboot. Once you've booted, you now run in the raid device (/dev/md0) Then you merely hook your original drive up to the raid array, it syncs and voila you're done.

So I set up my menu.lst to primarily load the kernel and initrd from the raid device, and failover to my original (still intact) old disk. And it always fails over when I reboot. I boot the machine, run my new grub entry and it says "error 15 file not found." Lots of stuff on the web about it, none seem to help.

The only thing that's weird is when I go to setup the MBR with grub, you say "root (hd0,0)" which I finally understand what it means, and it's supposed to say Filesystem type is ext2fs, partition type 0xfd or somethingn like that. Mine says nothing. But when I run setup (hd0) and setup (hd2) it says it's doing the right thing to the right drive. So I assume it's working. but it can't load initrd/the kernel from the md0 device.

The only other thing I'm thinking, is how on earth does grub know what a raid device is. The kernel hasn't loaded, the software raid modules haven't loaded, how can stupid little grub have any idea at all where to load initrd from? So I'm thinking, okay there's a mapping somewhere from /dev/md0 to /dev/sdc1 (the new raid drive) but I don't see where that could be happening. And for kicks, (I did this SO many times in various combinations) I tried setting the grub menu.lst to try and load the initrd and kernel from root=/dev/sdc1 (my new drive) and it still says file not found. So either the grub mbr setup isn't working, or I'm missing something really simple.

Any ideas?

Here's some more info...

root@io:~# cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md0 : active raid1 sdc1[1]
      18771840 blocks [2/1] [_U]



root@io:~# fdisk -l

Disk /dev/sda: 20.8 GB, 20847697920 bytes
255 heads, 63 sectors/track, 2534 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x9d949d94

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1   *           1        2337    18771921   83  Linux
/dev/sda2            2338        2434      779152+   5  Extended
/dev/sda5            2338        2434      779121   82  Linux swap / Solaris

Disk /dev/sdb: 320.0 GB, 320072933376 bytes
16 heads, 63 sectors/track, 620181 cylinders
Units = cylinders of 1008 * 512 = 516096 bytes
Disk identifier: 0x00000000

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb1   *           1        4064     2048224+  83  Linux
/dev/sdb2            4065      620181   310522968   83  Linux

Disk /dev/sdc: 20.0 GB, 20020396032 bytes
255 heads, 63 sectors/track, 2434 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00000080

   Device Boot      Start         End      Blocks   Id  System
/dev/sdc1   *           1        2337    18771921   fd  Linux raid autodetect
/dev/sdc2            2338        2434      779152+   5  Extended
/dev/sdc5            2338        2434      779121   82  Linux swap / Solaris

Disk /dev/md0: 19.2 GB, 19222364160 bytes
2 heads, 4 sectors/track, 4692960 cylinders
Units = cylinders of 8 * 512 = 4096 bytes
Disk identifier: 0x00000000

Disk /dev/md0 doesn't contain a valid partition table



root@io:~# mdadm -E
mdadm: No devices to examine



root@io:~# cat /etc/mdadm.conf
ARRAY /dev/md0 level=raid1 num-devices=2 UUID=5248ed76:cba39cc2:3082255a:649c0d18
root@io:~#



root@io:~# cat /boot/grub/menu.lst

default         0
# 8/14/09 added this
fallback        1

## timeout sec
# Set a timeout, in SEC seconds, before automatically booting the default entry
# (normally the first entry defined).
timeout         3

## hiddenmenu
# Hides the menu by default (press ESC to see the menu)
hiddenmenu

# added this 8/14/09 for raid boot, note this will get blown away on next kernel update
# if it's after the magic marker
# this means we will have to manually update this when there's a kernel upgrade :-(
# in grub land hd0 = /dev/sda and hd1 = /dev/sdb and hd2 = /dev/sdc I hope
# we're putting sdc first for now
title           Ubuntu 8.04.3 LTS, kernel 2.6.24-24-generic (raid)
root            (hd2,0)
#kernel         /boot/vmlinuz-2.6.24-24-generic root=UUID=b11d6b08-fdfe-4b0d-adec-4e263455be23 ro
kernel          /boot/vmlinuz-2.6.24-24-generic root=/dev/md0 ro
initrd          /boot/initrd.img-2.6.24-24-generic
quiet




title           Ubuntu 8.04.3 LTS, kernel 2.6.24-24-generic
root            (hd0,0)
kernel          /boot/vmlinuz-2.6.24-24-generic root=UUID=d8c402cc-7445-4878-b3aa-c9568b740b51 ro
initrd          /boot/initrd.img-2.6.24-24-generic
quiet


title           Ubuntu 8.04.3 LTS, kernel 2.6.24-24-generic (recovery mode)
root            (hd0,0)
kernel          /boot/vmlinuz-2.6.24-24-generic root=UUID=d8c402cc-7445-4878-b3aa-c9568b740b51 ro single
initrd          /boot/initrd.img-2.6.24-24-generic



root@io:~# blkid
/dev/sda1: UUID="d8c402cc-7445-4878-b3aa-c9568b740b51" SEC_TYPE="ext2" TYPE="ext3"
/dev/sda5: TYPE="swap" UUID="e0509276-30eb-4dcb-8e17-20f8244f5403"
/dev/sdb1: LABEL="alt" UUID="ea1789eb-9d6f-47a9-a074-18121792b30a" SEC_TYPE="ext2" TYPE="ext3"
/dev/sdb2: LABEL="sp" UUID="3b6d1173-f9fd-4a3e-8e5d-249fc682355b" SEC_TYPE="ext2" TYPE="ext3"
/dev/sdc1: UUID="76ed4852-c29c-a3cb-5a25-8230180d9c64" TYPE="mdraid"
/dev/md0: UUID="b11d6b08-fdfe-4b0d-adec-4e263455be23" SEC_TYPE="ext2" TYPE="ext3"

I've got a linux box that's running out of space, on a non-root volume drive. What's the best way to move to a larger drive?

I figure I boot to single user mode, format, mount the new drive, do some magical copy command that preserves links, permissions, file dates and everything else, then unmount the old drive, mount the new drive as the old name and reboot.

Does that sound right? Am I missing something? Suggestions? Tips? Anybody know what the cp command would be?

This is a ubuntu machine.

when I try and scp to my solaris box, I get this

unable to initialise SFTP: could not connect

unless I do scp -scp then it works.

In my sshd_config file, I have this:

Subsystem sftp /usr/local/libexec/sftp-server