Hi Simone, About single slave, when a slave goes down the master has a critical. I'd expect for all services on that host to become stale.
About the user/groups you have a permission editor, and may create partial admins (admins who can admin only a slave, and/or reload), read only users, ... The permissions are based on hostgroups, i.e. you can have a user only see a group of hosts; the access will always be via central server. Luis Periquito Operações [email protected] -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Simone Felici Sent: segunda-feira, 22 de Fevereiro de 2010 11:24 To: [email protected] Subject: Re: [opsview-users] Opsview: our next step for monitoring evolution? Hello Luis, Thank you for your answers! I rewrite the questions in hope someone can answer the opened questions: 1. Has someone tried import tool? Does it import historical data/logs/comments too? 2. OK for managed hosts/services. Other experiences would be nice. 3. Slaves config: Luis, you have answerred me about single slave. Someone experience with other question: Multi slaves in cluster. They are handled, as a separate datacenter, but the checks in this slave cluster are divided between all slaves in load balancing as well as in high availability. If one slave dies, the others take the services/hosts to be monitored. It's true? 4. What happens if single slave fails? I mean a slave with no clusters in group. All checks should be report critical. right? (should be) 5. OK for master acts as a slave. Question closed :) 6. OK for plugins too - I'll add a "-h" option in case I'll migrate to OPSView. New question: 7. Are there implementations of users and groups to give access only to a part of monitored devices? In case of a slave within a datacenter, I would like some customers can see only a part of the datacenter in RO, without see everything. Til today, creating manually a slave with only few hosts, the customer logs into remote slave and see only the few hosts. The master is accessable only by us. Thank's a lot for every answer! simon Luis F. Periquito ha scritto in data 22/02/2010 10.34: > Hi Simone, > > I'll try to answer some of your questions. To others I don't have answers... > > 1. Opsview community 3.5.1 is running on nagios 3.2. I don't know if the > import tools work as I never tried them. You should import configuration to > the Opsview, and then manage it from the GUI. You shouldn't make changes > directly in the nagios configuration as they will all be overwritten on the > next reload. Remember you have an integrated environment where all you should > do is on the GUI. > > 2. I currently have 296 hosts and 2746 services without breaking a sweat on > the master (1 QC xeon 2.33 with 4GB ram 32 bit debian etch) > > 3. > a) true. They will send the ncsa information back via an ssh tunnel. > b) never tried it. > > 4. never tried it. > > 5. the master shows up as it was a normal slave. > > 6. sure you can. Just have to make sure they print something when called with > -h option. > > Hope this helps. > > Luis Periquito > Operações > [email protected] > _______________________________________________ Opsview-users mailing list [email protected] http://lists.opsview.org/lists/listinfo/opsview-users _______________________________________________ Opsview-users mailing list [email protected] http://lists.opsview.org/lists/listinfo/opsview-users
