I want to upgrade our SAM environment to 6.0.0, while not upgrading any other environments. I am taking over this responsibility from another team.
1) Does the SAM upgrade require a stoppage of services? I realize that there are separate installs for Main, Poller & Web-only. I was curious if I needed to stop services before running the upgrade on the Main node first.
2) Is there anything else that I need to do besides running the install? Main -> Poller -> Web would be the order I go in.
3) Does the upgrade require a reboot?
Thanks - and sorry for the dumb questions.
Have you upgraded to SAM 6.0 yet ? We recently upgraded SAM 5.0.1 -> 5.5 (in order to be able to upgrade to NPM 10.6). Our SAM upgrade took 8hrs to run (with Orion completely down, no polling, no web server access). We do have considerably large tables, our database is 130GB, with APMHardware tables being quite large. Solarwinds did confirm that SAM upgrade can take considerable time. I'm curious what version you are on now, whether you have completed your SAM 6.0 upgrade, and if so did you experience a LONG upgrade?
epenney, an 8 hour upgrade is extremely unusual. If you'd like to determine why the upgrade is taking so long, I suggest opening a case with support. We'd be interested in looking a diagnostic before and after the upgrade to give us a better idea where the slowdown is occurring in your installation. Another option would be to consolidate your detailed data into hourly data before the upgrade, then increasing it again once the upgrade has completed. We believe this should significantly improve the speed of the upgrade, and reduce your down time.
|•||Change SAM's detail retention period to minimum (1 day). [Settings -> SAM Settings -> Data and Database Settings]|
|•||Run DB maintenance [Start -> Programs -> SolarWinds Orion -> Advanced Features -> Database Maintenance]|
|•||Increase SAM's retention period back to default value after upgrade.|
appreciate you taking time to provide input. I had opened a case during our extremely long SAM upgrade. It was support who told me that there were other customers who had experienced same, and to wait it out. Support did mention that there were specific SAM db/table changes as a result of going 5.0 -> 5.5 that were major contributing factors. I am now planning our SAM 5.5 -> 6.0 upgrade, and organizational management are asking me if I can reassure them that same outage will not repeat. As per your suggestion I altered our SAM data retention from 7 days detail / 30 days hourly / 90 days daily TO now read 1 day / 7 days / 30 days respectively. I also took a snapshot of current database tables (row counts & sizes) to see effect when next nightly runs. I am aiming for SAM 6.0 upgrade on Dec 11th, but only if I'm able to reassure management we won't experience same outage. Our orion platform manages network for majority of British Columbia, our usual maintenance window of 2hrs has been working out very well, until that last SAM upgrade
There are additional database table changes in the SAM 6.0 upgrade that will require migration, but reducing the amount of daily detail records in the database should significantly improve the upgrade data migration process performance. Keep us posted and let us know how it turned out. Also be sure to grab a diagnostic prior to upgrading in the event you encounter any issues, This will be the simplest way for support and development to know things like table sizes and index sizes before vs after the upgrade.
The upgrade will stop and start them automatically, but I always stop them in Orion service manager as a precaution. I also check after the upgrade to make sure they all are running.
SolarWinds solutions are rooted in our deep connection to our user base in the THWACK® online community. More than 150,000 members are here to solve problems, share technology and best practices, and directly contribute to our product development process. Learn more today by joining now.