Another Mid-tier cache question
Hello, Susan: The JSS SSO Plugin records user access including client type, browser, remote IP, and length of session time. More generally, every six months or so, I tend to add my two pence to the current Mid Tier cache thread, which is that the current cache model is not optimum. This is how I think it currently works: 1. Workflow (forms, active links etc) is loaded into local caches within Mid Tier using ehcache (google). It is persisted on disk and also loaded into memory. 2. When a request is made to load a form, HTML and accompanying Javascript is generated each time from the cached objects. There are a number of problems with this approach, not least that loading workflow into memory as Java objects is incredibly memory and CPU intensive, hence why Mid Tier and ITSM requires huge heap sizes and some Mid Tiers can take minutes to start. Also doing it each time Mid Tier starts means there's excessive consumption of AR System resources, in particular when people flush the cache rather than picking off workflow related to a particular form. The output of this process is an HTML and Javascript file, and hence there is a better way: 1. User makes request for /arsys/forms/appserver/Form. 2. Mid Tier requries the Javascript file to drive the workflow. Therefore, it looks for a file called appserver-Form.js n disk and serves if present. 3. If not, it walks the Form for all related workflow using a workflow analyser, loads from AR System and creates the Javascript file. Some will be cached, lots will be thrown away. 4. Mid Tier serves newly created Javascript file. 5. Workflow can be held in memory or ejected at this point. Most will not be needed again. (Repeat for local cached HTML.) There are of course some complications, ie some workflow will require loading on demand, but this avoids gigabytes of it being loaded into memory and converted into Javascript on every access to a form. On subsequent requests, Mid Tier is doing little more than serving the HTML and Javascript from local cached files, which incidentally, can easily be deleted by the administrator There are also other benefits. This is how 'developer' mode works: 1. User makes request for /arsys/forms/appserver/Form. 2. Mid Tier checks last modified timestamp on appserver-Form.js. 3. If timestamp different to the timestamp cached with appserver-Form.js then it deletes appserver-Form.js and re-creates. If not, file is served. So even the overhead of running in a always rebuild if changed mode is only a dent on performance. There's good news! I've mostly hacked up a solution to the above, but there's some Mid Tier internal code I can't access/modify. However, if I can sit down and make a reasonable attempt at resolving the problem in a few hours, surely someone in BMC could create a proof of concept for 'community testing'? If someone at BMC wishes to work with me to resolve it, I'm happy to do the coding for free :) John -- JSS SSO Plugin for the BMC product set http://www.javasystemsolutions.com/jss/ssoplugin ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
Re: Another Mid-tier cache question
Hey guys, Not sure if you are aware, but at the Evening with Engineers night at the WWRUG 2013, we were informed that there is a fix for cache corruption in the latest hotfix for MidTier 8.0 (not sure if it is solved for 8.1 or 7.6.04). This is the kind of corruption where fields/labels/etc. are in weird spots. I just downloaded it and will be testing/QA'ing in my environments soon. Thanks! -Sharon From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Jason Miller Sent: Friday, October 11, 2013 2:52 PM To: arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question ** The sync shouldn't have any affect on logging in. Even if the server isn't finished syncing yet the login process shouldn't care. If syncing was as hard on a system as a cache rebuild then I could see timeout issues occurring. You could probably tail the Mid Tier logs to see what is happening during a sync (depending on your logging settings). I have tailed the log when flushing cache but not syncing cache. The only way I know of seeing if a person is using MT or WUT is either 1) using workflow to capture it (in a record) or 2) watching access logs on the web server. I guess one other kind of goofy way is to tail the aruser.log. Since 7.6.04 or so when a user is using WUT there is a ton of logins recorded vs. when a user is using MT. If you see the same person logging in every few seconds then they are using WUT. On our environment where WUT is primarily used the aruser.log grows very rapidly. I choose always allow as well for pop-up blockers. Basically you want to trust any page MT is going to throw at you. It has been so long since I have read the message that I don't even remember what it say. Clicking the always allow happens from muscle memory at this point :) Jason On Fri, Oct 11, 2013 at 7:24 AM, Susan Palmer suzanpal...@gmail.commailto:suzanpal...@gmail.com wrote: ** I tried sync cache last night and it seemed a bit faster but still took at least 10-15 min. I have noticed that after a sync I have issues logging back into mid-tier which I assume means the sync is not done. Is there a place to watch what it's doing? Different question: Can you see who is logged on via the mid-tier as opposed to client tool. License review just shows everyone. Different question: Pop-up blockers. A bit confusing, it says 'don't have pop-ups active, then later it say it wants pop-ups. I actually had misread the 'don't allow' message wrong initially and had chosen 'always allow' and didn't see any ill effects of that. Thanks, Susan On Fri, Oct 11, 2013 at 9:12 AM, Brittain, Mark mbritt...@navisite.commailto:mbritt...@navisite.com wrote: ** Sync cache is disabled/grayed out so the only option I have is Flush Cache. From: Action Request System discussion list(ARSList) [arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG] On Behalf Of Joe D'Souza [jdso...@shyle.netmailto:jdso...@shyle.net] Sent: Thursday, October 10, 2013 5:13 PM To: arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question ** Like LJ said its best left off on production as your changes to Production if your company follows a proper change process is minimal. And you Sync the cache whenever there is a genuine change in the cache. Joe From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG] On Behalf Of ravi rai Sent: Thursday, October 10, 2013 5:07 PM To: arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question LJ, We do Quarterly releases.After each release we do manual flush Cache . Is is safe to turn this option off. It might resolve cache corruption issues which we encounter almost every alternate week Ravi Date: Thu, 10 Oct 2013 14:57:48 -0600 From: lj.longw...@gmail.commailto:lj.longw...@gmail.com Subject: Re: Another Mid-tier cache question To: arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.netmailto:jdso...@shyle.net wrote: ** I won't pretend to answer this question for you - but this is my guess.. From what it looks like, this functionality performs a periodic check on the AR Server
Re: Another Mid-tier cache question
Sharon, Thanks I did not think it was released yet. When I spoke to them, they said it was still in testing. hbr From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Perkins, Sharon (NTT Data) Sent: Monday, October 14, 2013 3:12 PM To: arslist@ARSLIST.ORG Subject: Re: [arslist] Another Mid-tier cache question ** Hey guys, Not sure if you are aware, but at the Evening with Engineers night at the WWRUG 2013, we were informed that there is a fix for cache corruption in the latest hotfix for MidTier 8.0 (not sure if it is solved for 8.1 or 7.6.04). This is the kind of corruption where fields/labels/etc. are in weird spots. I just downloaded it and will be testing/QA'ing in my environments soon. Thanks! -Sharon From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Jason Miller Sent: Friday, October 11, 2013 2:52 PM To: arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question ** The sync shouldn't have any affect on logging in. Even if the server isn't finished syncing yet the login process shouldn't care. If syncing was as hard on a system as a cache rebuild then I could see timeout issues occurring. You could probably tail the Mid Tier logs to see what is happening during a sync (depending on your logging settings). I have tailed the log when flushing cache but not syncing cache. The only way I know of seeing if a person is using MT or WUT is either 1) using workflow to capture it (in a record) or 2) watching access logs on the web server. I guess one other kind of goofy way is to tail the aruser.log. Since 7.6.04 or so when a user is using WUT there is a ton of logins recorded vs. when a user is using MT. If you see the same person logging in every few seconds then they are using WUT. On our environment where WUT is primarily used the aruser.log grows very rapidly. I choose always allow as well for pop-up blockers. Basically you want to trust any page MT is going to throw at you. It has been so long since I have read the message that I don't even remember what it say. Clicking the always allow happens from muscle memory at this point :) Jason On Fri, Oct 11, 2013 at 7:24 AM, Susan Palmer suzanpal...@gmail.commailto:suzanpal...@gmail.com wrote: ** I tried sync cache last night and it seemed a bit faster but still took at least 10-15 min. I have noticed that after a sync I have issues logging back into mid-tier which I assume means the sync is not done. Is there a place to watch what it's doing? Different question: Can you see who is logged on via the mid-tier as opposed to client tool. License review just shows everyone. Different question: Pop-up blockers. A bit confusing, it says 'don't have pop-ups active, then later it say it wants pop-ups. I actually had misread the 'don't allow' message wrong initially and had chosen 'always allow' and didn't see any ill effects of that. Thanks, Susan On Fri, Oct 11, 2013 at 9:12 AM, Brittain, Mark mbritt...@navisite.commailto:mbritt...@navisite.com wrote: ** Sync cache is disabled/grayed out so the only option I have is Flush Cache. From: Action Request System discussion list(ARSList) [arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG] On Behalf Of Joe D'Souza [jdso...@shyle.netmailto:jdso...@shyle.net] Sent: Thursday, October 10, 2013 5:13 PM To: arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question ** Like LJ said its best left off on production as your changes to Production if your company follows a proper change process is minimal. And you Sync the cache whenever there is a genuine change in the cache. Joe From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG] On Behalf Of ravi rai Sent: Thursday, October 10, 2013 5:07 PM To: arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question LJ, We do Quarterly releases.After each release we do manual flush Cache . Is is safe to turn this option off. It might resolve cache corruption issues which we encounter almost every alternate week Ravi Date: Thu, 10 Oct 2013 14:57:48 -0600 From: lj.longw...@gmail.commailto:lj.longw...@gmail.com Subject: Re: Another Mid-tier cache question To: arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer
Re: Another Mid-tier cache question
Sync cache is disabled/grayed out so the only option I have is Flush Cache. From: Action Request System discussion list(ARSList) [arslist@ARSLIST.ORG] On Behalf Of Joe D'Souza [jdso...@shyle.net] Sent: Thursday, October 10, 2013 5:13 PM To: arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question ** Like LJ said its best left off on production as your changes to Production if your company follows a proper change process is minimal. And you Sync the cache whenever there is a genuine change in the cache. Joe From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of ravi rai Sent: Thursday, October 10, 2013 5:07 PM To: arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question LJ, We do Quarterly releases.After each release we do manual flush Cache . Is is safe to turn this option off. It might resolve cache corruption issues which we encounter almost every alternate week Ravi Date: Thu, 10 Oct 2013 14:57:48 -0600 From: lj.longw...@gmail.com Subject: Re: Another Mid-tier cache question To: arslist@ARSLIST.ORG ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.netmailto:jdso...@shyle.net wrote: ** I won’t pretend to answer this question for you – but this is my guess.. From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. Cheers Joe From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG] On Behalf Of Brittain, Mark Sent: Thursday, October 10, 2013 4:34 PM To: arslist@ARSLIST.ORGmailto:arslist@ARSLIST.ORG Subject: Another Mid-tier cache question Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ This e-mail is the property of NaviSite, Inc. It is intended only for the person or entity to which it is addressed and may contain information that is privileged, confidential, or otherwise protected from disclosure. Distribution or copying of this e-mail, or the information contained herein, to anyone other than the intended recipient is prohibited. This E-mail and any of its attachments may contain Time Warner Cable proprietary information, which is privileged, confidential, or subject
Re: Another Mid-tier cache question
I tried sync cache last night and it seemed a bit faster but still took at least 10-15 min. I have noticed that after a sync I have issues logging back into mid-tier which I assume means the sync is not done. Is there a place to watch what it's doing? Different question: Can you see who is logged on via the mid-tier as opposed to client tool. License review just shows everyone. Different question: Pop-up blockers. A bit confusing, it says 'don't have pop-ups active, then later it say it wants pop-ups. I actually had misread the 'don't allow' message wrong initially and had chosen 'always allow' and didn't see any ill effects of that. Thanks, Susan On Fri, Oct 11, 2013 at 9:12 AM, Brittain, Mark mbritt...@navisite.comwrote: ** Sync cache is disabled/grayed out so the only option I have is Flush Cache. -- *From:* Action Request System discussion list(ARSList) [ arslist@ARSLIST.ORG] On Behalf Of Joe D'Souza [jdso...@shyle.net] *Sent:* Thursday, October 10, 2013 5:13 PM *To:* arslist@ARSLIST.ORG *Subject:* Re: Another Mid-tier cache question ** Like LJ said its best left off on production as your changes to Production if your company follows a proper change process is minimal. And you Sync the cache whenever there is a genuine change in the cache. Joe -- *From:* Action Request System discussion list(ARSList) [mailto: arslist@ARSLIST.ORG] *On Behalf Of *ravi rai *Sent:* Thursday, October 10, 2013 5:07 PM *To:* arslist@ARSLIST.ORG *Subject:* Re: Another Mid-tier cache question LJ, We do Quarterly releases.After each release we do manual flush Cache . Is is safe to turn this option off. It might resolve cache corruption issues which we encounter almost every alternate week Ravi -- Date: Thu, 10 Oct 2013 14:57:48 -0600 From: lj.longw...@gmail.com Subject: Re: Another Mid-tier cache question To: arslist@ARSLIST.ORG ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.net wrote: ** I won’t pretend to answer this question for you – but this is my guess.. From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. Cheers Joe -- *From:* Action Request System discussion list(ARSList) [mailto: arslist@ARSLIST.ORG] *On Behalf Of *Brittain, Mark *Sent:* Thursday, October 10, 2013 4:34 PM *To:* arslist@ARSLIST.ORG *Subject:* Another Mid-tier cache question Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where
Re: Another Mid-tier cache question
The sync shouldn't have any affect on logging in. Even if the server isn't finished syncing yet the login process shouldn't care. If syncing was as hard on a system as a cache rebuild then I could see timeout issues occurring. You could probably tail the Mid Tier logs to see what is happening during a sync (depending on your logging settings). I have tailed the log when flushing cache but not syncing cache. The only way I know of seeing if a person is using MT or WUT is either 1) using workflow to capture it (in a record) or 2) watching access logs on the web server. I guess one other kind of goofy way is to tail the aruser.log. Since 7.6.04 or so when a user is using WUT there is a ton of logins recorded vs. when a user is using MT. If you see the same person logging in every few seconds then they are using WUT. On our environment where WUT is primarily used the aruser.log grows very rapidly. I choose always allow as well for pop-up blockers. Basically you want to trust any page MT is going to throw at you. It has been so long since I have read the message that I don't even remember what it say. Clicking the always allow happens from muscle memory at this point :) Jason On Fri, Oct 11, 2013 at 7:24 AM, Susan Palmer suzanpal...@gmail.com wrote: ** I tried sync cache last night and it seemed a bit faster but still took at least 10-15 min. I have noticed that after a sync I have issues logging back into mid-tier which I assume means the sync is not done. Is there a place to watch what it's doing? Different question: Can you see who is logged on via the mid-tier as opposed to client tool. License review just shows everyone. Different question: Pop-up blockers. A bit confusing, it says 'don't have pop-ups active, then later it say it wants pop-ups. I actually had misread the 'don't allow' message wrong initially and had chosen 'always allow' and didn't see any ill effects of that. Thanks, Susan On Fri, Oct 11, 2013 at 9:12 AM, Brittain, Mark mbritt...@navisite.comwrote: ** Sync cache is disabled/grayed out so the only option I have is Flush Cache. -- *From:* Action Request System discussion list(ARSList) [ arslist@ARSLIST.ORG] On Behalf Of Joe D'Souza [jdso...@shyle.net] *Sent:* Thursday, October 10, 2013 5:13 PM *To:* arslist@ARSLIST.ORG *Subject:* Re: Another Mid-tier cache question ** Like LJ said its best left off on production as your changes to Production if your company follows a proper change process is minimal. And you Sync the cache whenever there is a genuine change in the cache. Joe -- *From:* Action Request System discussion list(ARSList) [mailto: arslist@ARSLIST.ORG] *On Behalf Of *ravi rai *Sent:* Thursday, October 10, 2013 5:07 PM *To:* arslist@ARSLIST.ORG *Subject:* Re: Another Mid-tier cache question LJ, We do Quarterly releases.After each release we do manual flush Cache . Is is safe to turn this option off. It might resolve cache corruption issues which we encounter almost every alternate week Ravi -- Date: Thu, 10 Oct 2013 14:57:48 -0600 From: lj.longw...@gmail.com Subject: Re: Another Mid-tier cache question To: arslist@ARSLIST.ORG ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.net wrote: ** I won’t pretend to answer this question for you – but this is my guess.. From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid
Another Mid-tier cache question
Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark This E-mail and any of its attachments may contain Time Warner Cable proprietary information, which is privileged, confidential, or subject to copyright belonging to Time Warner Cable. This E-mail is intended solely for the use of the individual or entity to which it is addressed. If you are not the intended recipient of this E-mail, you are hereby notified that any dissemination, distribution, copying, or action taken in relation to the contents of and attachments to this E-mail is strictly prohibited and may be unlawful. If you have received this E-mail in error, please notify the sender immediately and permanently delete the original and any copy of this E-mail and any printout. ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
Re: Another Mid-tier cache question
I won't pretend to answer this question for you - but this is my guess.. From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. Cheers Joe _ From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Brittain, Mark Sent: Thursday, October 10, 2013 4:34 PM To: arslist@ARSLIST.ORG Subject: Another Mid-tier cache question Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
Re: Another Mid-tier cache question
Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.net wrote: ** I won’t pretend to answer this question for you – but this is my guess..** ** ** ** From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. ** ** So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. ** ** I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. ** ** Cheers ** ** Joe -- *From:* Action Request System discussion list(ARSList) [mailto: arslist@ARSLIST.ORG] *On Behalf Of *Brittain, Mark *Sent:* Thursday, October 10, 2013 4:34 PM *To:* arslist@ARSLIST.ORG *Subject:* Another Mid-tier cache question ** ** Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
Re: Another Mid-tier cache question
LJ,We do Quarterly releases.After each release we do manual flush Cache .Is is safe to turn this option off.It might resolve cache corruption issues which we encounter almost every alternate week Ravi Date: Thu, 10 Oct 2013 14:57:48 -0600 From: lj.longw...@gmail.com Subject: Re: Another Mid-tier cache question To: arslist@ARSLIST.ORG ** Mark,I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.net wrote: ** I won’t pretend to answer this question for you – but this is my guess.. From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. Cheers Joe From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Brittain, Mark Sent: Thursday, October 10, 2013 4:34 PM To: arslist@ARSLIST.ORG Subject: Another Mid-tier cache question Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
Re: Another Mid-tier cache question
Based on that scenario, and my understanding of how things work...yes, you don't need to have the automatic check done periodically. I can't speak to the cache corruption, but one could surmise that if the cache is 'good', and you aren't updating it, ever, then there would be no possibility of it being corrupted:) On Thu, Oct 10, 2013 at 3:06 PM, ravi rai ravira...@hotmail.com wrote: ** LJ, We do Quarterly releases.After each release we do manual flush Cache . Is is safe to turn this option off. It might resolve cache corruption issues which we encounter almost every alternate week Ravi -- Date: Thu, 10 Oct 2013 14:57:48 -0600 From: lj.longw...@gmail.com Subject: Re: Another Mid-tier cache question To: arslist@ARSLIST.ORG ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.net wrote: ** I won’t pretend to answer this question for you – but this is my guess..** ** ** ** From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. ** ** So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. ** ** I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. ** ** Cheers ** ** Joe -- *From:* Action Request System discussion list(ARSList) [mailto: arslist@ARSLIST.ORG] *On Behalf Of *Brittain, Mark *Sent:* Thursday, October 10, 2013 4:34 PM *To:* arslist@ARSLIST.ORG *Subject:* Another Mid-tier cache question ** ** Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
Re: Another Mid-tier cache question
Like LJ said its best left off on production as your changes to Production if your company follows a proper change process is minimal. And you Sync the cache whenever there is a genuine change in the cache. Joe _ From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of ravi rai Sent: Thursday, October 10, 2013 5:07 PM To: arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question LJ, We do Quarterly releases.After each release we do manual flush Cache . Is is safe to turn this option off. It might resolve cache corruption issues which we encounter almost every alternate week Ravi _ Date: Thu, 10 Oct 2013 14:57:48 -0600 From: lj.longw...@gmail.com Subject: Re: Another Mid-tier cache question To: arslist@ARSLIST.ORG ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.net wrote: ** I won't pretend to answer this question for you - but this is my guess.. From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. Cheers Joe _ From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Brittain, Mark Sent: Thursday, October 10, 2013 4:34 PM To: arslist@ARSLIST.ORG Subject: Another Mid-tier cache question Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
Re: Another Mid-tier cache question
It would be nice if this option came with a indicator of what time will the next flush happen so that the mid-tier administrator would have a better visible control of what to set as an interval when setting this in non production environments. Joe _ From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of LJ LongWing Sent: Thursday, October 10, 2013 4:58 PM To: arslist@ARSLIST.ORG Subject: Re: Another Mid-tier cache question ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.net wrote: ** I won't pretend to answer this question for you - but this is my guess.. From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. Cheers Joe _ From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Brittain, Mark Sent: Thursday, October 10, 2013 4:34 PM To: arslist@ARSLIST.ORG Subject: Another Mid-tier cache question Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
Re: Another Mid-tier cache question
That would be nice. On Thu, Oct 10, 2013 at 2:15 PM, Joe D'Souza jdso...@shyle.net wrote: ** It would be nice if this option came with a indicator of what time will the next flush happen so that the mid-tier administrator would have a better visible control of what to set as an interval when setting this in non production environments. ** ** Joe ** ** -- *From:* Action Request System discussion list(ARSList) [mailto: arslist@ARSLIST.ORG] *On Behalf Of *LJ LongWing *Sent:* Thursday, October 10, 2013 4:58 PM *To:* arslist@ARSLIST.ORG *Subject:* Re: Another Mid-tier cache question ** ** ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. ** ** Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) ** ** On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.net wrote:*** * ** I won’t pretend to answer this question for you – but this is my guess..** ** From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. Cheers Joe -- *From:* Action Request System discussion list(ARSList) [mailto: arslist@ARSLIST.ORG] *On Behalf Of *Brittain, Mark *Sent:* Thursday, October 10, 2013 4:34 PM *To:* arslist@ARSLIST.ORG *Subject:* Another Mid-tier cache question Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ ** ** _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
Re: Another Mid-tier cache question
Our non-ITSM production MT is set to check every 60 minutes. That environment tends to have more changes done since it is a custom environment. The MT useage is lightish on this system since most people are still using WUT with this environment. Our ITSM 8 system is set to check every 10 hours. There are times we'll implement a minor change that is so minor or infrequently used functionality that I'll let the cache update on it own. This mentality is really from before Sync Cache was available when we didn't want to affect users by flushing cache. Jason On Thu, Oct 10, 2013 at 1:57 PM, LJ LongWing lj.longw...@gmail.com wrote: ** Mark, I agree with Joebut look at it this waythis check box tells the Mid-Tier server to periodically check your Remedy server for definition changes. How often are definition changes made in your production server?Weekly? Monthly? Quarterly? You likely don't need an automated 'check' to be turned on in production as it doesn't change very often...and when it does, you can manually hit the 'sync' button. Regarding the app server being behind a load balancer...no, that won't affect things because regardless of which app node the mid-tier gets the cache from, it should be 'correct' :) On Thu, Oct 10, 2013 at 2:45 PM, Joe D'Souza jdso...@shyle.net wrote: ** I won’t pretend to answer this question for you – but this is my guess..* *** ** ** From what it looks like, this functionality performs a periodic check on the AR Server, to check for changes in definitions, and collects that information. This will in my opinion have some impact on performance. ** ** So as long as that interval is relatively high, and set in such a way that it occurs in periodic cycles when users are usually not online, it should be fine. My guess is that when this box is checked and the interval is defined, there is probably a definition check that happens that instant, followed next by the interval that is defined. So if this is done lets say at 11:00 PM when most users are usually offline in that time zone, and the interval is set for 86400 for the next check to happen at 11:00 PM the next night, you might not have too much to worry about. ** ** I would however not be comfortable doing it every few minutes, as it MAY impact the performance of that particular mid-tier server in that load balanced configuration.. ** ** Cheers ** ** Joe -- *From:* Action Request System discussion list(ARSList) [mailto: arslist@ARSLIST.ORG] *On Behalf Of *Brittain, Mark *Sent:* Thursday, October 10, 2013 4:34 PM *To:* arslist@ARSLIST.ORG *Subject:* Another Mid-tier cache question ** ** Hi All, Is it safe to use Definition Change Check (Peform Check) with load balancers? When the dev and production ITSM servers were installed Perform Check was no selected. Don't know why it was done that way. Later when I applied a patch to the mid-tier servers, BMC Support said I should select Perform Check. I did this on the development server which has one ar server, one mid-tier and no load balancers, but did not select Perform Check on production which is a VIP load balanced to two mid-tiers which are load balanced to two ars servers in a server group. Particularly with small changes I really like using change check/perform check on dev and would like to use on the production servers. Since I don't know why this was not originally set up that way I figured I would ask the group first. ARS 7.6.06 SP3 Mid-Tier 7.6.06 SP4 Thanks Mark _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ _ARSlist: Where the Answers Are and have been for 20 years_ ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Where the Answers Are, and have been for 20 years
