Sysadmin Todo List

This is an unordered set of tasks. Detailed information on any of the tasks typically goes in related topics' pages, although usually not until the task has been filed under Completed.

Daily Check off list

Each day when you come in check the following:

1. Do not update nss_ldap on RHEL5 machines until they fix it
2. Einstein (script):
   1. Up and running?
   2. Disks are at less than 90% full?
   3. Mail system OK? (spamassasin, amavisd, ...)
3. Temperature OK? No water blown into room?
4. Systems up: Taro, Pepper, Pumpkin/Corn ?
5. Backups:
   1. Did backup succeed?
   2. Does Lentil need a new disk?

Important

• The einstein switchover seems to be roughly completed at first glance. Mail works, home directories work (tested with feynman, mariecurie, symanzik, lentil, pumpkin, and pauli) but some initially have a stale nfs file handle issue, which can be solved by a simple reboot. Blackbody seems to have weird permissions issues, but no one uses it right now, so it should be fine, as long as no other workstations act like it. DNS stuff is iffy at best. I haven't had the chance to make sure the cron stuff is all set nicely, but we seem operational! Now it's sleepytime.

Towards a stable setup

Here are some thoughts, especially to Steve, about getting an über-stable setup for the servers.
Some observations:

1. When we get to DeMeritt at the end of next summer, we need a setup that easily ports to the new environment. We will also be limited to a total of 10 kWatts heat load (36000 BTUs, or 3 tons of cooling), due to the cooling of the room. That sounds like a lot, but Silas and Jiang-Ming will also put servers in this space. Our footprint should be no more than 3 to 4 kWatts of heat load.
2. Virtual systems seems like the way to go. However, our experience with Xen is that it does not lead to highly portable VMs.
3. VMware Server is now a free product. They make money consulting and selling fancy add-ons. I have good experience with VMware Workstation on Mac's and Linux. But it is possible (like RedHat which was once free) that they will start charging when they reach 90% or more market share.

Here are some options:

• We get rid of Tomato, Jalapeno, Gourd and Okra and perhaps also Roentgen. If we want we can scavenge the parts from Tomato & Jalapeno (plus old einstein) for a toy system, or we park these systems in the corner. I don't want to waste time on them. The only bumps that I can think of here would be that Xemed/Aaron use Gourd. Otherwise I think we're all in favor of cutting down on the number of physical machines that we've got running. Oh, and what about the paulis? Since they're not under our "jurisdiction", they'll probably end up there anyhow.
• Test VMware server (See VMWare Progress). Specifically, I would like to know:

1. 1. How easy is it to move a VM from one hardware to another? (Can you simply move the disks?) Yes.
   2. Specifically, if you need to service some hardware, can you move the host to other hardware with little down time? (Clearly not for large disk arrays, like pumpkin, but that is storage, not hosts). Considering portability of disks/files, the downtime is the time it takes to move the image around and start up on another machine.
   3. Do we need a RedHat license for each VM or do we only need a license for the host, as with Xen? It seems to consume a license per VM. Following this didn't work for the VMWare systems. The closes thing to an official word that I could find was this.
   4. VMware allows for "virtual appliances", but how good are these really? Are these fast enough?

• Evaluate the hardware needs. Pumpkin, the new Einstein, Pepper, Taro and Lentil seem to be all sufficient quality and up to date. Do we need another HW? If so, what?

Einstein Upgrade

Einstein upgrade project and status page: Einstein Status Note: Einstein (current one) has a problem with / getting full occasionally. See Einstein#Special_Considerations_for_Einstein

Miscellaneous

• Learn how to use sieve a little better, and then send an email out to all users to let them know about it, and include an example to get rid of spam.
• Set up elog (looks like firewall is all that's left).
• Set up the new benfranklin.
• Fix some of the older workstations (parity, hobo, ennui, etc.)
• Set up a working java plugin on mariecurie. Also, figure out if there's a consistent way to get it working easily on 64-bit machines.
• Mariecurie: New NVIDIA video card seems to have the same problems as the ATI ones.
• Fermi has problems allowing me to log in. nsswitch.conf looks fine, getent passwd shows all the users like it's supposed to. There are no restrictions in /etc/security/access.conf, either.
• Gourd won't let me (Matt) log in, saying no such file or directory when trying to chdir to my home, and then it boots me off. Trying to log in as root from einstein is successful just long enough for it to tell me when the last login was, then boots me. (Steve here) I was able to log in and do stuff, but programs were intermittently slow.
• Monitoring: I would like to see the new temp-monitor integrated with Cacti, and fix some of the cacti capabilities, i.e. tie it in with the sensors output from pepper and taro (and tomato/einstein). Setup sensors on the corn/pumpkin. Have an intelligent way in which we are warned when conditions are too hot, a drive has failed, a system is down. I'm starting to get the hang of getting this sort of data via snmp. I wrote a perl script that pulls the temperature data from the environmental monitor, as well as some nice info from einstein. We SHOULD be able to integrate a rudimentary script like this into cacti or splunk, getting a bit closer to an all-in-one monitoring solution. It's in Matt's home directory, under code/npgmon/
• Check into smartd monitoring (and processing its output) on Pepper, Taro, Corn/Pumpkin, Einstein, Tomato.
• Decommission Okra. - This system is way too outdated to bother with it. Move Cacti to another system. Perhaps a VM, once we get that figured out?
• Learn how to use cacti. We should consider using a VM appliance to do this, so it's minimal configuration, and since okra's only purpose is to run cacti.

Ongoing

Documentation

• Maintain the Documentation of all systems!
  • Main function
  • Hardware
  • OS
  • Network
• Continue homogenizing the configurations of the machines.
• Improve documentation of mail software, specifically SpamAssassin, Cyrus, etc.

Maintenance

• Check e-mails to root every morning
• Check up on security [1]
• Clean up Room 202.
  • Start reorganizing things back into boxes for the August move.
  • Ask UNH if they have are willing/able to recycle/reuse the three CRTs and old machines that we have sitting around. Give them away if we have to. We're trying.

On-the-Side

• Learn how to use ssh-agent for task automation.
• Backup stuff: We need exclude filters on the backups. We need to plan and execute extensive tests before modifying the production backup program. Also, see if we can implement some sort of NFS user access. I've set up both filters and read-only snapshot access to backups at home. Uses what essentially amounts to a bash script version of the fancy perl thing we use now, only far less sophisticated. However, the filtering and user access uses a standard rsync exclude file (syntax in man page) and the user access is fairly obvious NFS read-only hosting. I am wondering if this is needed. The current scheme (ie the perl script) uses excludes by having a .rsync-filter is each of the directories where you want excluded contents. This has worked well. See ~maurik/tmp/.rsync-filter . The current script takes care of some important issues, like incomplete backups. Ah. So we need to get users to somehow keep that .rsync-filter file fairly updated. And to get them to use data to hold things, not home. Also, I wasn't suggesting we get rid of the perl script, I was saying that I've become familiar with a number of the things it does. [2]
• Continue purgin NIS from ancient workstations, and replacing with files. The following remain:
  • pauli nodes -- Low priority!

Waiting

• That guy's computer has a BIOS checksum error. Flashing the BIOS to the newest version succeeds, but doesn't fix the problem. No obvious mobo damage either. What happen? Who was that guy, anyhow? (Silviu Covrig, probably) The machine is gluon, according to him. Waiting on ASUS tech support for warranty info Aaron said it might be power-supply-related. Nope. Definitely not. Used a known good PSU and still got error, reflashed bios with it and still got error. Got RMA, sending out on wed. Waiting on ASUS to send us a working one! Called ASUS on 8/6, they said it's getting repaired right now. Wohoo! Got a notification that it shipped! ...they didn't fix it... Still has the EXACT same error it had when we shipped it to them. What should we do about this? I'm going to call them up and have a talk, considering looking at the details on their shipment reveals that they sent us a different motherboard, different serial number and everything but with the same problem.

Previous Months Completed

June 2007

July 2007

August 2007

September 2007

October 2007

NovDec 2007

January 2008

February 2008

March/April/May/June 2008 (I'm doing a great job keeping track of this, eh?)