Jas's questions

I have 30 grafana panels in a dashboard. If I add a new one to bottom and then drag it to top it replaces the panel which was in the top to be in the bottom which is totally what I don't want I want it to be inserted and the top panel to move to the right or left, is that possible?

I need to support few server names like:

ab1
ab2
ab3
abc

so the last letter can change I thought of having

server {
        listen 80;
        server_name ab?;
        return 200 "i'm the server";
}

is this correct? how can I test it?

thanks

I'm looking for a way to zgrep hdfs files

something like:

hadoop fs -zcat hdfs://myfile.gz | grep "hi"

or

hadoop fs -cat hdfs://myfile.gz | zgrep "hi"

it does not really work for me is there anyway to achieve that with command line?

I tried to make alias with quotes as following:

alias myalias='ps -ef | grep tomcat | kill -9 `awk {'print $2'}`'

but as you can see i already have ' in awk

so i tried to replace

awk {'print $2'}

with

awk {"print $2"}

but then strange things happen to me when i run this alias, ie, the console window get closed... how can i make this alias work

I'm running

top -H -b -n 1

on a Linux with a java app

I get result with headers:

PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND 

is the %CPU I see for each thread a real cpu meaning it does not take into account the time the thread wasted on waiting for locks, etc., so its purely the time the thread has been spending on CPU.

i'm on a ubuntu machine when i fail to install jenkins with puppet with the following error:

/etc/puppet/modules/jenkins/manifests# puppet agent --no-daemonize --verbose
notice: Starting Puppet client version 2.7.1
info: Caching catalog for fqdn.myhost.com
err: Could not run Puppet configuration client: Parameter unless failed: 'apt-key list | grep -Fqe 'D50582E6'' is not qualified and no path was specified. Please qualify the command or specify a path.

and indeed i don't have this:

/etc/puppet/modules/jenkins/manifests# apt-key list | grep -Fqe 'D50582E6'
/etc/puppet/modules/jenkins/manifests#

i have followed explanation in this page and use first module specified there:

https://wiki.jenkins-ci.org/display/JENKINS/Puppet

i have downloaded the apt module from: https://github.com/camptocamp/puppet-apt/downloads

and placed it in puppet under modules.

anyone can help how can i overcome this? thanks

I'm using mcollective & puppet for deployments. I would like to know if there is any user interface which allows me to choose a specific package/version and deploy and any user interface i can already use which allows me to show the deployments history and their status.

(I saw that glu has a nice user interface i'm looking for something like it, http://linkedin.github.com/glu/docs/latest/html/tutorial.html

)

I wondered if i could use glu (so that i can get their deployment history gui if they have a good one...) with mcollective but from what i understand they are parallel frameworks.

i'm starting to use puppet. i wanted to have my own yum repository to serve puppet server so i issued

 yum install createrepo

no errors reported (cenotos 5.5) after that i issued

createrepo /var/www/html/repository

but i get:

shell-init: error retrieving current directory: getcwd: cannot access parent directories: No such file or directory
Traceback (most recent call last):
  File "/usr/share/createrepo/genpkgmetadata.py", line 724, in ?
    main(sys.argv[1:])
  File "/usr/share/createrepo/genpkgmetadata.py", line 592, in main
    cmds, directories = parseArgs(args)
  File "/usr/share/createrepo/genpkgmetadata.py", line 431, in parseArgs
    cmds['basedir'] = os.getcwd()
 OSError: [Errno 2] No such file or directory

note the folder:

/var/www/html

exists

I configured pound like this:

ListenHTTP
        Address 192.168.21.10
        Port    1632
        Service
                BackEnd
                        Address 192.168.22.220
                        Port    1622
                End
                BackEnd
                        Address 192.168.22.220
                        Port    1612
                End
        End
End

ListenHTTP
        Address 192.168.21.10
        Port    1634
        Service
                BackEnd
                        Address 192.168.22.220
                        Port    1534
                End
                BackEnd
                        Address 192.168.22.220
                        Port    1644
                End
        End
End

I do see with tcpflow it is getting requests on192.168.21.10 on port 1632 but its not forwarding to my server which i have listening on 192.168.22.220 (and not forwarding the requests (not proxying) to any other server...

its my first time using pound i know its up by its service, i don't see any log messages on var/log/messages

can anyone help me please? thanks

I intend to use chef or puppet to do administration (I'm thinking more of chef as it's younger and I get a better feeling about it).

In both home pages I saw there is an "enterprise edition" that costs money and I don't intend to buy anything. What would I miss in chef / puppet if I don't buy them?

What does chef offer that costs money exactly?
What does puppet offer that costs money exactly?

It was not so clear to me from their web site, as it's kind of obscure.

I'm interested in juju before starting with it i would like to ask a few questions.

1. Can juju be only be used with ubuntu? is it possible to use it with CentOS?
2. I have my own application server which i wrote, can i wrap it in a charm and deploy it on my local juju cloud? (I do not publish ofcourse my charm to the openstack because its a propriety server code, if yes can someone direct me maybe to a link of how i can achieve such a thing?) my server is built with java.
3. Is ubuntu server + full capabilities of juju always be free of charge?
4. Which technologies would you recommend me to write my app server installer (i'm going to go for continuous deployment on my own cloud rolled, perhaps with juju).

How can i achive the following with HAProxy with regards to its simple tutorial found here:

http://www.webhostingtalk.com/showthread.php?t=627783

I would to use HAProxy first direct all trafic to primary if off direct to secondary if primary back redirect back to primary

Is there also an option to discard some of the requests by the proxy? for example all requests /tothisplace to discard at this HAProxy level?

Also is there anything much simpler than HAProxy that I can use to achieve that? (I used in the past simpleproxy for other usage and it was very easy to use, is there a way to do it with simpleproxy?)

thanks

I have a java app that accesses an NFS share with hard mount (soft mount does troubles i get many RPC timeout's for some reason). I'm on solaris. If i turn off (svcadm -v disable to the nfs server) then my java app gets stuck if doing

kill -3 myPID

returns nothing as my java process hangs - strange, why even kill -3 to get thread dump does not return? also i even had problems killing my process - its not something i want to do, i don't want my app to get stuck if the mount is down Anyone has a solution or can recommend what I should be doing? I would like to use maybe soft mounts the problem is that i get many RPC timeout while I try to copy files from one server to an nfs share.

 NFS write failed for server xx.xx.xx.xx: error 5 (RPC: Timed out)
 cp: /share/myfile.txt: Connection timed out

so because i'm getting these errors with soft mounts and it looks unreliable (its over tcp) then i wished to test hard mount but this makes my app stuck if the share is down for some reason.

I want it to stop monitoring it for sometime so that I can stop my server (my main motivation is that smf wont start my server again automatically once i manually stopped it) and then start it again (i will start manually) and then tell smf to start monitoring it again... thanks

after installing monit when i do monit status myproc i get "error connecting to the monit daemon"

I read somewhere that

The status command won't work in the case that monit is running indaemon mode without its http support - the command 'monit status' in such case tries to get the status from the daemon via http/tcp. To start the http interface you need to add the 'set httpd ...' statement to theconfiguration.

is that still correct? that post was from 2005

I'm interested to access large log files on a unix server with textpad. (textpad for history reasons, i personally prefer ofcourse less awk grep etc) but I have many personal who rather be using textpad they have years of experience with it and can tweak it to do whatever they want.

The problem is that if i connect for example with winscp to get the log files to textpad it first fetches the full log and user needs to wait and it bloats etc. I would rather the textpad to somehow access the unix machine and get only the relevant segment of the log file (large log files could be GB) anyone knows how can this be achieved?

Currently I have windows only environment with multiple servers and some server access shared directories in other servers like this

\server1\sharedir1 (win server) \server2\sharedir2 (win server) \server3\sharedir3 (win server)

1. if i convert server 3 to be a solaris 10 and I install SAMBA on it will it be able because of sanmba installation to continue accessing \server1\sharedir1?
2. will the windows machines be able to access because of samba the shared dir \server3\sharedir3? because I have samba on it?
3. do i need to install something more on my windows 2003 server?

Thanks!

I already have an SVN server set up on a windows server machine (got it by inheritance) on this same machine I also have got cruisecontrol running.

First i'm going to separate cruisecontrol from SVN - SVN is so important I'm not going to let cruisecontrol run on this very same server!!!

My questions:

1. I plan to migrate from windows (couple of years hardware) to centos, do I need new hardware? (my sysadmin recommends that - less serverfaults... more support)
2. What are Minimal Hardware requirements for svn server + cruisecontrol on secondary vm?

So my plan is this:

1. Leave the SVN server on windows server for beginning.
2. Move cruisecontrol into a virtual server, what should be the minimal requirements? (I want it to be centos) of the hardware like? 1GB memory? 100GB disk? what else?
3. On that very same virtual machine put a back of the svn server + data in case the main svn server has an error - standby SVN server.
4. Get a new machine for SVN server (do I need a new machine?) or can I get away with an older one? (my sysadmin told me that we need to buy a new machine as the hardware is old - a couple of years, more prone to failures and also less support). on that new machine install centos and install the svn server that will be the main svn server.

Anyone has any recommendations for me? Thanks

which os should I choose solaris or linux if linux which distribution? i want to deploy my java app on production servers in a cluster on tens of servers... can someone tell me what are the key points I should compare between the 2 os? which os should I choose and why?

I'm using a remote desktop inside another machine to connect to another machine with another remote desktop (even a couple of more times of such a chain)

however when i'm inside the second one (or thirdone in chain) and i want to get back to its parent i cant do it.

when i go to top of the screen the title bars of all remote desktops overlay one on each other, even if i hit the clip button on the remote desktop title it does not work all of them goes up.

i tried shortcuts like ctrl-alt-break nothing works i searched for all keyboard shortcuts for remote desktop and nothing help...

can anyone help me please?