Erik I's questions

Sometimes I'd like an easy way to produce lists over everything a server accepts or connects to. Not what it can connect to/accept connections from (firewall rules etc) but actual connections.

I can kind of solve this in a number of ways:

• running ss or netstat on a schedule and collect data (how often and how long to make sure nothing slips through? yuck)
• get firewall to log almost everything (not always an option)

Both these ideas should allow me to (with some postprocessing and additional data gathering to ) to generate lists of

• which processes
• send and/or receive data (ok, maybe technically just connects in some cases)
• on what ports
• to or from which addresses

But I still have a feeling I might be missing a simpler solution. Does something like inotify / Filesystemwatcher exist for networks? Or something that can be used to similar effect? I'm more or less comfortable in both Powershell/Bash and a few mainstream programming languages so it is OK even if it requires some assembly.

This question here is kind of similar but the tools recommended seems to be geared more towards what can be connected to, not where data actually flows: Automated Network Mapping

Zabbix agent complains when trying to start:

Starting Zabbix agent: zabbix_agentd [...]: ERROR: cannot add user parameter 
"mysql.status[*],echo "show global status where Variable_name='$1';" |
HOME=/var/lib/zabbix mysql -N | awk '{print $$2}'": key "mysql.status" 
already exists

The file has been copied almost exactly from the existing userparameter_mysql.conf which has been renamed.

Looking to deploy an existing Mongodb-backed solution to the cloud for a client.

I'm hoping to combine MongoHQ and Azure to use the simplest possible Mongodb deployment together with the simplest possible .Net deployment.

Possible problems I can think of on top of my head includes latency (how much to expect from crossing between clouds) and extra bandwidth usage (this of course depends completely on the app in question).

My question is if anyone here has done anything similar before and knows any pitfalls I should be aware of?

Background:

On an almost freshly installed pc I get a message along the lines of : "windows cannot find some-file-server-name. Check the spelling and try again"... when trying to access any fileshare.

Troubleshooting so far:

• pinging works. Both by ip and by name
• the almost identical pc next to this one can access the file server
• everyone else can access the file server
• the pc in question can not access other open fileshares
• but it can connect to the internet

And now for what I think is the interesting part:

• running wireshark with ip.addr == local.ip.add.ress and ip.addr == server.ip.add.ress tells me that it tries to connect over http.
• the server replies but after a few messages back and forth it stops
• the other machine of course just uses smb

I guess port 80 just means it defaults to webdav, but I haven't been able to find anything that can cause this.

Googling it the closest thing I found was this http://www.techrepublic.com/article/get-vista-and-samba-to-work/6353849 but then again this was an XP pc and I wasn't able to connect to other native Windows shares (and I tried the solution anyway and it didn't work.)

I want to find out if there is a CD drive on a remote server running Red Hat Linux release 7.3 (Valhalla).

Based on what I can see in fstab, /proc/devices, /proc/ide/ide0/ it seems like there is a CD drive, but is there any way I can verify it?

[root@someserver yum]# yum update
Loaded plugins: rhnplugin, security
rhel-x86_64-server-5                                     | 1.4 kB     00:00     
rhel-x86_64-server-5/primary                             | 3.4 MB     00:03     
rhel-x86_64-server-5                                                10025/10025
rhel-x86_64-server-supplementary-5                       | 1.4 kB     00:00     
rhel-x86_64-server-supplementary-5/primary               | 250 kB     00:00     
rhel-x86_64-server-supplementary-5/primary               | 250 kB     00:00     
rhel-x86_64-server-supplementary-5/primary               | 250 kB     00:00     
rhel-x86_64-server-supplementary-5/primary               | 250 kB     00:00     
Error: failed to retrieve repodata/f0************************************************74-primary.xml.gz from rhel-x86_64-server-supplementary-5
error was [Errno -1] Metadata file does not match checksum
[root@someserver yum]#

google says:

• https://access.redhat.com/kb/docs/DOC-25842
• https://bugzilla.redhat.com/show_bug.cgi?id=480183
• https://www.redhat.com/archives/redhat-list/2009-January/msg00101.html

Based on the information above I've tried

yum clean all
rm -rf /var/cache/yum/*

in addtion to enabling and disabling the different "channels".

Anybody knows the solution to this?

I am running into a problem where I have created some ASM disks under linux, they show up when I run oracleasm listdisks, but they are invisible when I use dbca and try to Configure Automatic Storage Management.

Server is HP DL380 running Red Hat 4.

Server is cloned by swapping drives in raid. When starting in new server, kudzu finds 4 nics removed, 4 nics added, and adds default ifcfg-eth0 - 4 in /etc/sysconfig/network-scripts/

Then I basically tried to follow this procedure http://www.cyberciti.biz/tips/linux-bond-or-team-multiple-network-interfaces-nic-into-single-interface.html to bond the interfaces.

The server it is cloned from runs fine with bonded nics.

Anyone have a good idea of what might be wrong or how to troubleshoot it?

I have a linux server which needs to be set up with production ip address and hostname before installation procedures start (some scripts use these variables).

I've seen somebody assigning a temporary ip address and a temporary gateway simply by typing a couple of commands, but I can't remember.

Reason is I want to make a tidy script instead of

cd /some/dir
cpio -whatever<somefile
cd -