Re: [Vserver] strange things

2004-06-07 Thread Herbert Poetzl 
On Mon, Jun 07, 2004 at 06:20:24AM +0200, Dimitri Roschkowski wrote:
 Hi,
 
 my last Mail was about the crash of the master server by a high load, 
 that cause by some vserver. Yesterday I found a realy strange thing, 
 which happend one minute befor the master server crashed:
 
  11:17:01 up 1 day, 22 min,  0 users,  load average: 3.17, 2.57, 2.37
 USER TTYLOGIN@   IDLE   JCPU   PCPU WHAT
 CTX  PROCVSZRSS  userTIME   sysTIMEUPTIME NAME DESCRIPTION
 0  34  117MB3kB   2h04m27  28m36.45   1d12h22 root server
 49163   361GB   13kBm08.32m02.97   1d12h19 v12
 49166   542GB   52kB   4m56.48   4m33.37   1d12h19 v15
 49169   15   39MB3kB   2m29.66   1m22.73   1d12h18 v18
 49170   12  158MB2kBm49.51m11.68   1d12h18 v19
 49174   41  647MB   21kB   5m19.56   1m02.99  11h16m34 v10
 49175   33  489MB   20kB   3m29.60m40.07   6h59m42 v01
 49176   40  555MB   24kB   3m26.92m41.36   6h59m22 v02
 49177   33  484MB   19kB   3m21.93m37.78   6h58m54 v03
 49178   33  484MB   18kB   3m19.57m38.10   6h58m33 v04
 49179   46  795MB   50kB   5m29.05   1m48.18   6h58m20 v05
 49180   39  504MB   20kB   3m10.16m56.96   6h58m08 v06
 49181   38  506MB   20kB   3m16.94m38.14   6h57m48 v07
 49182   33  483MB   18kB   3m18.24m38.33   6h57m36 v08
 49183   31  474MB   17kB   3m18.96m37.63   6h57m24 v09
 49184   19  111MB3kBm02.93m00.50   6h57m24
 49185   23  179MB6kBm03.54m00.79   6h57m07
 49186   29  429MB   13kB   2m15.02m28.12   6h57m07 v11
 49187   29  427MB   12kBm01.51m00.80   6h56m55
 49188   23  140MB4kB   1m24.48m25.44   6h56m54 v13
 49189   27  455MB   16kB   2m13.71m29.22   6h56m35 v14
 49191   32  475MB   17kB   2m14.63m27.78   6h56m16 v16
 49192   26  154MB6kB   2m11.79m28.77   6h56m10
 49194   26  454MB   14kBm03.47m01.16   6h55m55 v20
 
 As you can see, there are 24 vservers running (of 20 vservers). Whats 
 the Problem? This happend one minute before the master  server (suse 9.0 
 with 2.4.25 and vs1.26) crashed.
 
 Another thing I noticed about SuSE vservers running on a SuSE master 
 server, that the vserver can take over the local terminal, if this lines 
 are  enabled in the /etc/inittab of a vserver:
 1:2345:respawn:/sbin/mingetty --noclear tty1
 2:2345:respawn:/sbin/mingetty tty2
 3:2345:respawn:/sbin/mingetty tty3
 4:2345:respawn:/sbin/mingetty tty4
 5:2345:respawn:/sbin/mingetty tty5
 6:2345:respawn:/sbin/mingetty tty6
 
 Thats my config file:
 IPROOT=eth0:
 IPROOTDEV=eth0
 ONBOOT=yes
 S_HOSTNAME=x
 S_DOMAINNAME=x
 S_NICE=
 ULIMIT=-HS -u 80
 S_CAPS=CAP_NET_RAW
 S_FLAGS=lock nproc fakeinit
 
 How can I prevent it?

hmm, I'd try to update to 2.4.26 + vs1.27 just because
I 'think' that some kernels issues (between 2.4.25 and
2.4.26) might be the reason, otherwise I'd check the
hardware, as linux-vserver is _very_ stable in the 
stable branch ...

best,
Herbert

 Cheers
 Dimitri
 
 ___
 Vserver mailing list
 [EMAIL PROTECTED]
 http://list.linux-vserver.org/mailman/listinfo/vserver
___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver

Re: [Vserver] VServer management

2004-06-07 Thread Dennis Roos 
On 4 Jun 2004 at 18:39, Lucas Albers wrote:

 
 Dariush Pietrzak said:
   Hmm,  there is another issue here - if you already use app like HP
   Open
  View to do your other management, then putting it in control of
  vservers might be the wisest choice.
 
  --
 What would be some useful cluster commands?
A couple of examples might be:
 * VServer replication
 * load-balancing applications
 
 What are some common vserver operations?
Status checking
IP management (depending on the underlying management)
Disk usage checks
Package updates
Services checks
Server creation/removal
Password resets for users
...
 
 Move vserver from one machine to another.
 Determine if any vservers are stopped.
 Stop,start,create vservers.
 
 The most useful imo is the replication.
 Steps:
 rsync vserver from one machine to another.
 copy over conf files.
 stop the old vserver.
 replicate the data.
 start the new vserver.
 Currently I do this by hand, if it was scripted I would have less
 chance of screwing it up, and shorter downtime on the move.
I'm looking into ways of automating these steps, hence the question 
on SNMP or alternate ways to do remote management. I prefer a 
trigger-reply mechanism to accomplish these tasks.

However, there is an easier way to better the uptime, using the lvs 
way. The only issue here is keeping 2 vservers online and in-sync 
the latter is probably the hardest part ;)


Regards,
Dennis Roos

Network Engineer
InTouch N.V.
Middenweg 76
1097 BS Amsterdam
Tel: +31 (0)20 6752060
Fax: +31 (0)20 6758429


___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver

Re: [Vserver] network interface in vs 1.9.1

2004-06-07 Thread Bert De Vuyst 
On Thursday 03 June 2004 21:07, Enrico Scholz wrote:
 [EMAIL PROTECTED] (Bert De Vuyst) writes:
  I have a test setup running vs 1.9.1 on a stock 2.6.6 kernel, using
  util-vserver 0.29.214.
  To get the networking running in the vservers,

 Please define running.

Sorry, I was a bit confused.
Running the stable versies of linux-vserver, I did check the network settings 
using ifconfig. Running the development versies, this doesn't seem to work.
But when I create the file name in the /etc/vservers/VSERVER-NAME/
interfaces/0/ (and give the interface a name), the interface does show op in 
ifconfig.

 Use 'ip addr' to display the available IPs.
You are wright, ip addr gives the running IP-numbers. And the networking is 
running (apache does work)

Thanks for your help,

Bert.
___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver