Accessible method of measuring the impact of recycling an app pool and / or restarting a service.
Hello, Do any of you have a method of testing the impact in terms of CPU, Disk and RAM usage when restarting an IIS app pool or the W3W service / a process within this service? I need to form an argument against restarting an app pool every hour. I know why this isn't a good idea but I need to quantify it. Thanks
Hi, We use an external website uptime facility and whenever the app pool is recycled the site response time goes from a few ms to over 30 seconds on some sites. Even though other users are accessing the site, all of them have to wait until the site is fully working and that includes the external uptime monitor. In fact we have an automatic failover facility on that system which actually engaged one time we did an app pool recycle. I've been there with the app pool refresh every hour and the way I got round it was to stress that session data was lost on restart, meaning users in our case were kicked back to the login screen. I still have some web services where I need to periodically recycle if specific changes are made as certain records are only read when the app pool comes up. This is no lie, I have a script to restart a console app every hour for licensing reasons! I would love to come in the office to find it down because of a restart failure, unfortunately I haven't been lucky yet :). Andrew. -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Darragh Ó Héiligh Sent: 27 January 2015 16:51 To: blind-sysadmins@lists.hodgsonfamily.org Subject: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hello, Do any of you have a method of testing the impact in terms of CPU, Disk and RAM usage when restarting an IIS app pool or the W3W service / a process within this service? I need to form an argument against restarting an app pool every hour. I know why this isn't a good idea but I need to quantify it. Thanks _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins
Hi, We use an external website uptime facility and whenever the app pool is recycled the site response time goes from a few ms to over 30 seconds on some sites. Even though other users are accessing the site, all of them have to wait until the site is fully working and that includes the external uptime monitor. In fact we have an automatic failover facility on that system which actually engaged one time we did an app pool recycle. I've been there with the app pool refresh every hour and the way I got round it was to stress that session data was lost on restart, meaning users in our case were kicked back to the login screen. I still have some web services where I need to periodically recycle if specific changes are made as certain records are only read when the app pool comes up. This is no lie, I have a script to restart a console app every hour for licensing reasons! I would love to come in the office to find it down because of a restart failure, unfortunately I haven't been lucky yet :). Andrew. -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Darragh Ó Héiligh Sent: 27 January 2015 16:51 To: blind-sysadmins@lists.hodgsonfamily.org Subject: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hello, Do any of you have a method of testing the impact in terms of CPU, Disk and RAM usage when restarting an IIS app pool or the W3W service / a process within this service? I need to form an argument against restarting an app pool every hour. I know why this isn't a good idea but I need to quantify it. Thanks _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins
Andrew, what's the monitoring tool? -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Andrew Hodgson Sent: 27 January 2015 18:31 To: Blind sysadmins list Subject: Re: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hi, We use an external website uptime facility and whenever the app pool is recycled the site response time goes from a few ms to over 30 seconds on some sites. Even though other users are accessing the site, all of them have to wait until the site is fully working and that includes the external uptime monitor. In fact we have an automatic failover facility on that system which actually engaged one time we did an app pool recycle. I've been there with the app pool refresh every hour and the way I got round it was to stress that session data was lost on restart, meaning users in our case were kicked back to the login screen. I still have some web services where I need to periodically recycle if specific changes are made as certain records are only read when the app pool comes up. This is no lie, I have a script to restart a console app every hour for licensing reasons! I would love to come in the office to find it down because of a restart failure, unfortunately I haven't been lucky yet :). Andrew. -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Darragh Ó Héiligh Sent: 27 January 2015 16:51 To: blind-sysadmins@lists.hodgsonfamily.org Subject: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hello, Do any of you have a method of testing the impact in terms of CPU, Disk and RAM usage when restarting an IIS app pool or the W3W service / a process within this service? I need to form an argument against restarting an app pool every hour. I know why this isn't a good idea but I need to quantify it. Thanks _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins
Hi, Sorry, its part of the services that Dyn offer with their managed DNS service. It works by measuring the response time to get a specific HTTP/S response - in our case containing a specific set of keywords from the login screen of our application. We also used AppDynamics which was external and also usable with JFW, but it was ditched in favour of a ManageEngine product due to cost. This Applications Manager product doesn't seem to observe the latency in recycling the app pool. There are countless other monitoring products out there with external/cloud versions which you could trial. Andrew. -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Darragh Ó Héiligh Sent: 27 January 2015 18:31 To: Blind sysadmins list Subject: Re: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Andrew, what's the monitoring tool? -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Andrew Hodgson Sent: 27 January 2015 18:31 To: Blind sysadmins list Subject: Re: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hi, We use an external website uptime facility and whenever the app pool is recycled the site response time goes from a few ms to over 30 seconds on some sites. Even though other users are accessing the site, all of them have to wait until the site is fully working and that includes the external uptime monitor. In fact we have an automatic failover facility on that system which actually engaged one time we did an app pool recycle. I've been there with the app pool refresh every hour and the way I got round it was to stress that session data was lost on restart, meaning users in our case were kicked back to the login screen. I still have some web services where I need to periodically recycle if specific changes are made as certain records are only read when the app pool comes up. This is no lie, I have a script to restart a console app every hour for licensing reasons! I would love to come in the office to find it down because of a restart failure, unfortunately I haven't been lucky yet :). Andrew. -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Darragh Ó Héiligh Sent: 27 January 2015 16:51 To: blind-sysadmins@lists.hodgsonfamily.org Subject: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hello, Do any of you have a method of testing the impact in terms of CPU, Disk and RAM usage when restarting an IIS app pool or the W3W service / a process within this service? I need to form an argument against restarting an app pool every hour. I know why this isn't a good idea but I need to quantify it. Thanks _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins
Hi, Sorry, its part of the services that Dyn offer with their managed DNS service. It works by measuring the response time to get a specific HTTP/S response - in our case containing a specific set of keywords from the login screen of our application. We also used AppDynamics which was external and also usable with JFW, but it was ditched in favour of a ManageEngine product due to cost. This Applications Manager product doesn't seem to observe the latency in recycling the app pool. There are countless other monitoring products out there with external/cloud versions which you could trial. Andrew. -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Darragh Ó Héiligh Sent: 27 January 2015 18:31 To: Blind sysadmins list Subject: Re: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Andrew, what's the monitoring tool? -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Andrew Hodgson Sent: 27 January 2015 18:31 To: Blind sysadmins list Subject: Re: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hi, We use an external website uptime facility and whenever the app pool is recycled the site response time goes from a few ms to over 30 seconds on some sites. Even though other users are accessing the site, all of them have to wait until the site is fully working and that includes the external uptime monitor. In fact we have an automatic failover facility on that system which actually engaged one time we did an app pool recycle. I've been there with the app pool refresh every hour and the way I got round it was to stress that session data was lost on restart, meaning users in our case were kicked back to the login screen. I still have some web services where I need to periodically recycle if specific changes are made as certain records are only read when the app pool comes up. This is no lie, I have a script to restart a console app every hour for licensing reasons! I would love to come in the office to find it down because of a restart failure, unfortunately I haven't been lucky yet :). Andrew. -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Darragh Ó Héiligh Sent: 27 January 2015 16:51 To: blind-sysadmins@lists.hodgsonfamily.org Subject: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hello, Do any of you have a method of testing the impact in terms of CPU, Disk and RAM usage when restarting an IIS app pool or the W3W service / a process within this service? I need to form an argument against restarting an app pool every hour. I know why this isn't a good idea but I need to quantify it. Thanks _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins
Andrew, what's the monitoring tool? -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Andrew Hodgson Sent: 27 January 2015 18:31 To: Blind sysadmins list Subject: Re: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hi, We use an external website uptime facility and whenever the app pool is recycled the site response time goes from a few ms to over 30 seconds on some sites. Even though other users are accessing the site, all of them have to wait until the site is fully working and that includes the external uptime monitor. In fact we have an automatic failover facility on that system which actually engaged one time we did an app pool recycle. I've been there with the app pool refresh every hour and the way I got round it was to stress that session data was lost on restart, meaning users in our case were kicked back to the login screen. I still have some web services where I need to periodically recycle if specific changes are made as certain records are only read when the app pool comes up. This is no lie, I have a script to restart a console app every hour for licensing reasons! I would love to come in the office to find it down because of a restart failure, unfortunately I haven't been lucky yet :). Andrew. -----Original Message----- From: Blind-sysadmins [mailto:blind-sysadmins-bounces@lists.hodgsonfamily.org] On Behalf Of Darragh Ó Héiligh Sent: 27 January 2015 16:51 To: blind-sysadmins@lists.hodgsonfamily.org Subject: [Blind-sysadmins] Accessible method of measuring the impact of recycling an app pool and / or restarting a service. Hello, Do any of you have a method of testing the impact in terms of CPU, Disk and RAM usage when restarting an IIS app pool or the W3W service / a process within this service? I need to form an argument against restarting an app pool every hour. I know why this isn't a good idea but I need to quantify it. Thanks _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins _______________________________________________ Blind-sysadmins mailing list Blind-sysadmins@lists.hodgsonfamily.org https://lists.hodgsonfamily.org/listinfo/blind-sysadmins
participants (2)
-
Andrew Hodgson -
Darragh Ó Héiligh