Friday, February 22, 2008

Week in review

Hey readers, sorry I haven't been on top of things, but there is too much to keep on top of, and its hard to have time to dump all of this. Now my email has let up…. But the snow is still pouring down I have a few minutes to think and post my thoughts.

We had a great week of progress in my groups on various projects. Here are some of the accomplishments for this week that we had. I will try to post more info on the things we tackle:

  • HP NNM
    • Moved to APA polling, and alarms have been reduced quite a bit.
  • HP RAMS
    • Integration is now working, and defining the paths we want to alarms on.
    • Had some good training, and hopefully the product will start to deliver better value than we have gotten from it in the past
  • HP BAC
    • Still having issues getting events from BAC to flow into OVO for notifications from BPM/Sitescope
    • Implemented a lot of new monitors for up/down to replace some older Spectrum monitoring that was in one of the groups.
    • Looking at RUM in the future, might replace an existing tool and give us better functionality.
  • Sevone
    • Things are in much better shape now, and we are pushing out some additional ACL changes to fix access from one of the Sevone clusters.
    • The install in the UK is progressing well now.
    • The reporting requests are flooding in, so lots of work. We want to educate users to use it themselves, so we can stop being bothered with simple requests.
  • Splunk
    • Working on some smaller issues, the product is bulletproof
  • Opsware NAS
    • Working on 3 annoying/major bugs, making progress on them, support had been slow previously, but thanks to bothering my sales guy things have picked up a lot.
    • I have 2 fixes for some rendering bugs as well, which seem to be in the next patch.
  • Opsware PAS
    • Dev upgrade was botched, but working with support. Will do production once we figure out what the problem is.
  • Vmware Virtualcenter / Platespin
    • Making progress on access and integration. Expanding some of our SQL clusters to take the additional DBs.
  • IBM Director
    • Installed the power monitoring module, its quite nice, same story as everywhere. We have no power.
    • Working on the integration via SNMP into Netcool Omnibus.
  • Netcool
    • Worked on deduplication problems and made other changes. We've been losing events, and its been more problematic recently. This is a major thorn in our side.

Friday, February 1, 2008

Sevone and Other new products

We have 2 of our clusters up running Sevone, which is our replacement for some other network polling/graphing systems. The product is excellent, and easy to use. I love the peer to peer data exchange it uses to facilitate a redundant system of clusters. The product is a breeze to deal with. We have found a few bugs, but nothing too bad. The discovery has been running for a few days, and its going quite slowly.

We have another 1400 devices to discover J

Took a look at a few products the last 2 weeks:

  • CA DVSM – Why run this when we have to run virtualcenter too? We are a vmware shop.
  • Solidcore – product is great, but we are not mature enough for it IMHO. It's a good value add for SAS.
  • Sun Ops Center – too immature, and Sun hasn't done well in this space historically. The product roadmap looks like it will be nice in a couple yrs.
  • Firescope – product is too broad, I have tools that do 90% of the features, I just want a good dashboarding and consolidator that is EASY and SCALABLE. I already have one anyways… BAC J

Only have 1373 devices left now!

Virtualcenter

Getting ready to deploy the big virtualcenter infrastructure, which should be fun since I don't have the staff hired yet J

Just reviewing final plans which are going live in the next few weeks.

HP/Opsware SAS

Making good progress on a business which wants a SAS deployment and I'm working on a business case (I think this is the 5th one). It would be great for us and for them if it materializes. We'll see if it flies in the next few weeks.

NNM Redesign

HP has come into the messed up, half functional NNM 7.53 implementation we have and proposed some major changes to fix the issues which are plaguing us. The propose to move us to APA polling, but we need to also implement CDP consistently. I am already getting a lot of flack for wanting to implement CDP. Aside from this project knowing when we have duplex mismatches, which you can imagine we get a lot of would be nice for operations J

We saw some nice future direction regarding Opsware product, which we already use, but that's as far as I'm going to go! NNM 8 looks to be a very nice product; it's a matter of us waiting for some SPIs and a bit more maturity. I hope that it has it by late this year.

BAC/BPM/Sitescope

We are deploying monitoring on these products now. We are slowly moving customers off OVIS now onto these products. We have a consulting company which should be starting in a few weeks on the mostly manual conversion of the OVIS HTTP_TRANS probes over to BPM/Vugen scripts.

We have built a lot of dashboards and such now, and they are working very well so far. I am looking forward to getting it into full production. DR is going to happen in a couple months for BAC on a partially virtual infrastructure. The events are moving from BAC into OVO in order to be escalated. We haven't yet integrated any OVO data into BAC, and there doesn't seem to be documentation on doing that.

It's something I'm planning on getting into with HP in the next 6-8 weeks.