How did I get back here again?
I went VM (Proxmox FTW) and Ansible crazy. Somewhere in the VM build I included S@H. I was using the S@H client to test with. Well, I forgot until a ps -Aef on a work node and whaaat??? The client wasn't hurting anything and it all in my "dev" space so I just let them go.
Thoughts about SETI and SETI@home
1. Why do you run SETI@home?
Answer: Sparky's fault...... And it became a ansible testing point for me. That is, I just forgot to remove it when I started using it to build of VM worker hosts.
- name: Copy from NFS to local FS.
- name: Unpack SaH.
2. What are your views about the project?
Answer: I don't really have one. The </Some/Path/>setiathome.berkeley.edu/setiathome_8.00_x86_64-pc-linux-gnu binary makes for a good stress test for my worker VMs. As far as the data that is being parsed by the VM's I don't know. The resulting data is useful somewhere from what I understand.
3. Any suggestions?
Local RDB support for engines Oracle/Postgres/MySQL (Sorry I am not a big on Mircosoft SQL Server services for data warehousing, maybe ODBC support for MS peeps?) to allow the clients to pull Work units from a database as to local filesystem of $SetAtHome_base/BOINC/projects/setiathome.berkeley.edu. This would allow for all SETI at home parsers to work collectively on X work units and, in the event a worker node failed the the user account would not suffer the lost of all work units in $SetAtHome_base/BOINC/projects/setiathome.berkeley.edu for that node. I am sure some pre parseing could also happen to make things "faster". This has the "bonus" to allow Teams to "share" work units. Team 1 and Team CG both bind to SETI at Home DB Server L337! DB Server is hosted on UBERBBQPWN Datacenter. Team 1 have Great CPU power and stable InterNET. Team CG is all GPU and on dial up which is prone to fail. You get the idea....
I know this is very high level and all the nuts and bolts would need to get worked out in long, all day meetings with clueless execs, and bean counters, where no one with any technical experience was invited. Change Management, that more worried about start and end times, and do you have a back out plan, what is the justification and did you have the right groups, blah blah blah, and no one with any technical experience is in that group. Incident management team that does not understand "tar fvc /working/prod/directory /backup/SaveMyKeyster-$(date %F).tar" is a good idea before "rm -rf /working/prod/directory" because no one with any technical experience is in that group. Business owners scream that is my data you cant have it because the business owners do not talk to the technical experience in that group! Clueless access control teams that thinks root_squash is the same as chmod 0007 on NFS mount. "Owner bit is 0, it is squash" because no one with any technical experience is in that group. (Yes I really had that fight * Face desk *). All controlled by an upper MBA management layer of single, make up covered, cougars that take baths in perfumes, or smell like she just walked out of Sephora. All too busy flirting with the hot hunky high school male interns. Or worried about the next edition of Louis Vuitton red bottom shoes, to make her legs look good for the hot hunky high school male intern she is flirting with therefore forgetting to make a decision on if we can look into adding RDB support to SETI at home. And yes, no one in upper MBA management layer of leadership has any technical experience. Yes, all of this has happened to me at one point or another and in some cases more then once. Ugh!