1 of 1 people found this helpful
I've heard of the superman job description and this sounds like the same for this server- all this functionality and you can't have a seperate DB server- for how many nodes?
Given the severe lack of resources I assume you too will be constrained so getting as much as possible pre-integrated from vendors will help you. So here we go-
- Help Desk management
(Web Help Desk since its already integrated, is easily customizable, and includes some change management- however I think this would require another server)
- Network inventorying
(I use NPM/UDT for this more then NCM but either would work)
- Infrastructure availability monitoring (with alerts and reporting of up-times)
(NPM for network devices/power infrastructure and SAM for everything else)
- Load monitoring (with alerts and graphs for reporting)
(same and its best to rely on the top 10 summary for outliers and also add all your devices into functional groups ( or custom property such as "group" or "system") so you can see load monitoring across these like systems such as all your blackberry servers)
- Change management (for configurations, assets and procedures documentation)
("1% inspiration and 99% perspiration" - this is so much more about all levels of management supporting a process vs what tools are used to implement it. However assuming you have said support then it goes back to your constrained resources. With that in mind and in an attempt to Keep It Simple I would use WHD for end user and enginnering process CR (build up the ticket templates for consistency, require NPM/SAM auditing to be cut/paste into ticket to show it was implemented as requested, FAQ as needed, and then NCM for consistent network device configuration management. )
- Patch management
(I have not used Languard and just digging into Patch Manager but leveraging wsus while getting reporting in Solarwinds is a plus)
- Document management
(we have an entire sharepoint farm and team but it sounds like you could do a NAS as a doc store behind a wiki page or even google docs)
Hope that helps?
1 of 1 people found this helpful
We have a similar challenge at my company. One of our consultants called our situation "Tool Soup".
I hope you receive many more responses on your query!
I think that network inventory, availability and load monitoring can all be handled well with Solarwinds. The others don't fit so well.
Running all of this on just a single server might be not be possible or practical. For instance, Soalrwinds recommends to separate the SQL server from the Solarwinds / Orion server.
Sorry I didn't have the "Holy Grail" answer. Best of luck in your quest!
How big is this environment?
I will say that it sounds like a tall order - not impossible but I sense that underheath the hood is a management attitude of "anything is possible when you don't know what you are talking about".
Thanks for the advice. Now I am pushing for an ESXi server, with 128GB ram and 6 10k sas disks in raid 5 with nv memory controller. So I'll be able to run multiple systems, containing the costs.
Regarding document and systems change management do you have any advice?
I'm not a big fan of Sharepoint. I know, I know, it's a corporate standard. I hate it. Openly and with the burning fire of 1,000 suns.
I like setting up a wiki (pick any, they run on Linux and they are easy to install and maintain) for knowledgebase. Plus, they use regular URL's so you can actually link a KB article inside your Solarwinds alert as part of the response actions. Wiki's allow for file attachments so that
And for as for change management, that's so focused on people processes more than technology that almost anything COULD work. Even a simple open source system like RT (Request Tracker) that allows for workflows and groups.
Don't use RAID 5... not anymore at least. It's a disaster waiting to happen. RAID 5 was relevant when HDDs were small but now that they are big and somewhat inexpensive, you should always go with RAID 10. In RAID 5 with big arrays, when you lose a disk, the risk of URE (unrecoverable read error, which breaks the array) during rebuild is higher than the risk of losing a second disk (which also breaks the array).
thanks for the advise syldra, I am using 300GB HDDs, we are not going to run business critical systems on it, so for us raid 5 will be fine
Good, sorry to hijack your thread like this but I find it important to spread the word... I've seen too many RAID 5 systems with 2-3TB drives... rebuild takes forever and has a good chance of ruining your array.