Ron,
That does make sense to me. As Joe points out, the function dumps ALL cache and then reloads it. I am not certain but based upon what you found, the cache may not have been completely refreshed from the restart of the server. (Perhaps if I make enough steps out of bounds in answering you, Doug Mueller himself will step in and "correct" me, or at least fill in the blanks ) If I remember (a big IF), perhaps I will bring up this issue at the WWRUG on the night we meet with the Engineers (including Doug). If I don't remember, perhaps someone on the list will remind me when we are standing around the BMC Engineering staff (I heard Darius is going to be there this year - Again)! One of the interesting things about what you are doing is that even when you only suspect something may be wrong (like seeing a setting you would expect to be refreshed with a restart or cache flush) it is most helpful, if time allows (for non-production issues mainly) to capture - even if it is just a screen shot - what you find before you perform the function that you think may correct or otherwise impact the current conditions. As you have discovered, after the proposed fix is put in place it is often times impossible to identify the root cause. Taking data points along the way may seem like a lot of work to do for very little return, but the fact that you have theorized what you suspect may have been the cause is a good gut feel indication that you were probably right. The use of the resulting documentation can be used by the rest of the team and put into a KB or other repository that would otherwise go by the wayside. As a consultant, it is truly valuable for me to leave my customers with documentation of what was performed and why. Although this may appear to eliminate the need for them to hire you back, it actually works in your (and their) favor. Both parties benefit since they not only have the "work product" and supporting documentation of your efforts, but you instill a level of trust with all those you interact with. Good in case they want to call you back for more work as well as leaving a trail of great references JJ Phil Bautista, WWRUG11 Advisory Board 512-731-0304 Social - <http://www.facebook.com/phil.bautista> http://www.facebook.com/phil.bautista Business - <http://www.linkedin.com/in/philbautista> http://www.linkedin.com/in/philbautista WWRUG11 - <http://www.wwrug11.com/contact_phil.html> http://www.wwrug11.com/contact_phil.html From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Ron Tavares Sent: Wednesday, August 31, 2011 9:48 AM To: arslist@ARSLIST.ORG Subject: Re: arsignal -r in a server group ** Thanks for the info Phil. Interesting that you should say that "a much more efficient method to synchronize the servers". Because I believe the arsignal was successful at caching some settings that were out-of-sync, which a regular ar restart was not able to capture. Though I can't confirm that. But it was regarding the Menu Access settings in the view properties. Example, Search option was turned off, yet the search button appeared on the Incident form. This remained the case for about a year, ar server restarts never caught this, until I ran the arsignal. Or at least that is my theory. Does this make sense to you? .ron On Wed, Aug 31, 2011 at 9:58 AM, Phil Bautista <bauti...@bullcreek.com> wrote: ** Ron, Depending on the number of servers that you have in your server group, ten minutes is not bad. I have used this command in a small server group (two servers) for a SIT environment and depending on the number of objects that are required to be synchronized it rarely took less than ten minutes. For a production server group containing ten servers, it has taken significantly longer. While you may have experienced equal or lesser times to simply restart the BMC Remedy Arserver service there are other processes that are dependent on the service. I would take a look a the dependencies on the BMC Remedy Arserver service and perform some tests if you prefer to compare the two options and gather comparable times and verify the additional dependent functions to see if all of the services come up in the same time it takes you to simply run the arsignal command. I am betting it doesn't You should be able to execute this command from the primary admin server in the server group. There should not be a need to go to each server to perform the command. Bottom line, this typically a much more efficient method to synchronize the servers in your server group and should be capable of executing from your admin server. Ten minutes is exceptional for a production server group and even for some development and test server groups. Sounds like a great topic for a birds of a feather session at the upcoming WWRUG11 or a question to ask at the Evening with Engineering! Phil Bautista, WWRUG11 Advisory Board 512-731-0304 <http://www.linkedin.com/in/philbautista> http://www.linkedin.com/in/philbautista <http://www.wwrug11.com/contact_phil.html> http://www.wwrug11.com/contact_phil.html From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Ron Tavares Sent: Wednesday, August 31, 2011 5:40 AM To: arslist@ARSLIST.ORG Subject: arsignal -r in a server group ** Looking for some expertise on using this little gadget. I have been experimenting with using arsignal -r to recache the servers in our server group, as an alternate to restarting the ar system service on all the servers. I'm thinking this will reduce the impact to the end users. However, in my tests, executing this command hangs the server for about 10 minutes. Is this normal? Because for that matter, might as well just restart the service. Is there any advantace to using arsignal instead of a restart. Also, the documentation says to execute as follows arsignal -r <server_name> where server_name is the name of the server that is to reload the information, and that you can execute from any server. So, I figure I can execute it from the admin server, once for each server. But I find that this does not work, that I have to go to each server, and execute it from there. Is that correct? ARS 7.1 Thanks, .ron _attend WWRUG11 www.wwrug.com <http://www.wwrug.com/> ARSlist: "Where the Answers Are"_ _attend WWRUG11 www.wwrug.com <http://www.wwrug.com/> ARSlist: "Where the Answers Are"_ _attend WWRUG11 www.wwrug.com ARSlist: "Where the Answers Are"_ _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org attend wwrug11 www.wwrug.com ARSList: "Where the Answers Are"
