Mister IT Guru's questions

I'd like to use CoreOS as my base OS going forward.

I run a lot of apps in multiple data centers, and I fully believe that I should containerize the lot. This has thrown up a lot of questions, notably control and access of resources.

My dream - is to have a cluster that runs my apps across multiple hosts and scales as required. When I (or the team) wish to make any changes to the cluster, we set a flag or variable in etcd which should trigger a script that updates the cluster - which I feel is possible

My fear comes when I realise that we have a lot of none routable networks that where data is stored, and multiple locations across the world -- Am I going to have to make all these networks routable, if I am to utilize the distributed keychain?

If so, then this kills my dream of running a local instance of CoreOS and connecting it to the cluster, and having access to all the information without having to actually log into a production cluster member?

I hope this makes sense -- I guess I'd like to control my cluster by sending requests to an end point, rather than having to be locally present on the cluster to make the changes. This allows an easy integration path for our existing control scripts and automated systems, which I'd really do not want to have to rebuild all of that!

I've got a clean instance of CentOS and a clean instance of RHEL 6.3 - I've tried to installed Moodle on both.

Error: Database connection failed

It is possible that the database is overloaded or otherwise not running properly.

The site administrator should also check that the database details have been correctly specified in config.php

The database does work - I've tried it from a shell on the box. I've tried with a remote database, and with a local database.

I've tried to remove the config.php and clear the database, and allow moodle to create run it's fresh install script, and again, I have the same problem.

The stange thing is, that I have RHEL 5.x and 6.x boxes that are running moodle instances, but I never had a problem installing them, so I never really looked closly at the stack. What is worrying me, is that these failures are occurring on my new system images - Systems that have been updated to the latest available packages in the repo in the last few weeks.

Has anyone else come across a similar problem?

I have to setup a small windows network inside my bigger linux/mac infrastructure. In order to get the windows clients logging onto the domain, I have had to make the DC their primary DNS server, which seems to have worked.

I would much prefer to have one DNS server running on my network, or at least one authoritative server running on the network.

I have a USG 200 router/firewall and I can configure some static records for DNS, but I an not sure what I need to put in order to get DNS and AD working together, and hints and tips appreciated.

I'd like to know how to hide a user from the OSX login screen. I'd like to have a hidden administrator account for using ARD, rather than the user's main account.

Any hints, tips and ideas?

I'm attempting to connect Kerio to an Open Directory instance. I'm using kinit to test the setup, and I get the following:

$ kinit -V -S host/[email protected] [email protected]
Please enter the password for [email protected]:
Kerberos Login Failed: Cannot resolve network address for KDC in requested realm

This occurs, even though I am 100% sure that the password is correct. Either way, I am more concerned with regards to the part of the message that says

Kerberos Login Failed: Cannot resolve network address for KDC in requested realm

From the server running OD, and a second in house server I get the same error message. I can dig and ping server.domain.co.uk correctly from both servers, so it boggles my mind what could be wrong.

I need this to be working before I can move forward and connect up the Kerio instance to my OD.

edu.mit.kerberos

[libdefaults]
    default_realm = SERVER.domain.CO.UK
[realms]
    SERVER.domain.CO.UK = {
        admin_server = server.domain.co.uk
        kdc = server.domain.co.uk
    }
[domain_realm]
    domain.co.uk = SERVER.domain.CO.UK
    .domain.co.uk = SERVER.domain.CO.UK
[logging]
    admin_server = FILE:/var/log/krb5kdc/kadmin.log
    kdc = FILE:/var/log/krb5kdc/kdc.log

SERVER is the real hostname for the machine in question, and domain.co.uk is my FQDN, or at least replacing my FQDN

Thanks for any assistance.

I'm trying to get my sharepoint workspaces on office 365 to mount like network drives. I can achieve this following microsoft guidelines, but after a while this fails. The user has to login again to their 365 account, and then remount the drives.

This is time consuming and annoying and with 50+ users you can understand that would produce a bucket load of support calls.

I've given this issue to microsoft to deal with and they have explained that because of the use of form authentication, the drive mapping will time out after 2 hours. (I can't be bothered to find the email just yet - I'll add it to this question if it's requested.)

I'd like to know if others have suffered this issue, and found a workaround? Or is there a way to prevent this disconnection?

Any hints and tips appreciated.

How can import an existing linux machine into Rackspace is pretty much what I'd like to know. I've got a VM hosted in a datacenter, and I'd like to move that VM, which is configured and running perfectly to Rack Space Cloud Server

I don't really like the idea of having to use only rackspace images, but if I really have no choice then any ideas on how to upload my own image to rackspace? Or indeed any cloud provider.

Thanks,

Every now and again, (once a day for some users), emails get saved in the draft folder and not sent. There is no message to show this when it happens, just a few hours later, you notice unread messages in your drafts folder.

I'm trying to find out why this happens, as some of the messages cannot be sent by simply clicking opening them and clicking send.

Some assistance would be appreciated.

I'm currently running IMAPsync on CentOS and I'm getting out of memory errors. The problem I'm having is that I'm using the smallest mailbox as a test case, and it's not helping my confidence level to have out of memory errors on a 2G mailbox.

I assumed from the description on the page that is was a recursive program, so I expected it to continue from where it left off, but I have no idea if it did that on the second run or not.

Is it possible to have IMAPsync continue until both mailboxes are in sync, or at least till all the mail from host1 has been moved to host2?

Thanks

If I install the splunk forwarder, I can get the remote data into my splunk install, and index my logs, and searching is great. But I have a number of router devices and other devices that run syslog, and can export their logs somewhere.

How can I configure Splunk to receive these logs, or is there some other workaround to be used?

I have around 300 baby macs to look after, and they have lots of different admin accounts, which is a nightmare to work with.

ARD has slowly been configured with the accounts of the machine owners, but that doesn't help when bringing on board another sysadmin, who now has to collect all the usernames and passwords of the users.

I figure, it's better for me to run a script using ARD that creates a hidden admin user on each system, with a known username and password. I just don't know how to do it.

Any hints and tips would be appreciated. I am currently trying to work on a draft script now, which I'll post, but anyone else, feel free to jump in with a solution :)

Is it possible, (seeing as Apple Open Directory is LDAP), to manage the users names and passwords for Ubuntu and Fedora based systems?

I have a challenge presented before me, and that is to centralise authentication. Period.

That's because me and my big mouth said I like LDAP, and anything can authenticate against it. I have pretty much every type of desktop here, Macs, Windows XP up to 7, and number of Ubuntu and Fedora based distro installs.

I have no problem going through all the configuration work. In fact that should be pretty fun, I just want some recommendations on which implementations I should be looking at.

Thank You

I have a client in a managed office, and part of that service is the provision of internet, with static IPs for all desktops provided by the building, (I know it's a recipe to have all your data stolen, and you not even know!). They insist on saying that traffic is not managed, but we have different bandwidth available on different sockets.

Two desktops have blisteringly fast internet, and the rest have dodgy crappy internet access.

Is there anyway that I can prove that the traffic is being shaped, so that I can convince the powers that be to bring networking inhouse?

I have network based printers, and I was wondering, as all network printers with SNMP able to give up information such as if they need ink, or toners or just their general status.

I'm not the greatest fan of walking up and down flights of stairs checking printers, because users can't be bothered to mention that a printer isn't working. (Unless they have a deadline, in which case, they use the broken printer, and somehow remember exactly how to log incident tickets!)

Any hints and tips appreciated. Thanks

I need to retire a particular piece of hardware, I have another server online on this particular network.

I have run DCPromo om the new server, but when the original server is shutdown, authentication on the network fails. I have also made the new 2003 server a global catalogue, and authentication still fails.

I would like to know how I can 'retire' the old server. It may need to stay on the network for legacy reasons, but I would dearly love to remove all domain activity from it without any downtime.

I look forward to help in this regards! Thanks for your tips hints and comments.

All my bacula jobs are writing to one file. How can I migrate these jobs to separate files, I have a 1G limit. I've already marked the volume as full, and yet, the volume is still being used.

I have multiple windows 2003 servers support a particular network, and when one particular server is taken offline, domain services begin to rapidly fail, such as users couldn't login.

What check list shall I look into to ensure that I can take my primary domain controller offline, for work, or any other reason, and have the domain still work as expected. I thought I had redundancy, but actually, it turns out that I didn't.

I'd like to look into this and figure out how to actually have redundancy for the domain services - and hints, tips, and comments very much appreciated.

I have a group of users in AD which reside at DOMAIN\mediapeople and i can assign this group to a share, like i have done on to a file share on the domain controller.

I have a new server, which also hosts some large media files, (this is it's purpose). I wish to add this group DOMAIN\mediapeople to this new share, but it is not available? I have a RDP session open to the main server, and also this secondary server, logged in with the same credentials, and this security group is available to one server, but not the other This is the first I have ever seen this, and I am wondering if anyone else has ever seen this before?

I backup some of my windows servers about 3 times a day because of the large amount of data that changes on it.

I had a hardware failure that killed one of my machines, and now I have a problem. I can restore that server to a virtual machine no problem, using Acronis, and get it booting up, but it asked me to activate it. I click to activate online, yet it fails. I have the key, and it's a valid windows instance, but I fear I will not be able to get this system back online anytime soon. Microsoft are being no help either.

Is it possible to change the key to a VLK that we have, that we just didn't use on this server, because it came with windows 2003 on it already?