Default pool
Publish: 2021-04-29 17:02:47
1. Default pool in amoeba.xml means default pool
2. Application pool; DefaultAppPool' The shutdown time of the process providing services exceeds the limit
the server often generates "application pool"; DefaultAppPool' The shutdown time of the serving process exceeded the limit. The process ID is; 2068'” Is in the state of feign death. It is understood that the problem is the setting of IIS application pool. The solution is as follows:
Internet information services (IIS) manager > application pool > defaultapppool > right click Properties
1. Recycle work process (minutes): selected, value 1740
2. Recycle work process (number of requests): not selected (originally set to 35000)
3. Recycle work process at the following time: not filled
4 Recycle worker processes when consuming too much memory: none 2. 3 and 4 items may avoid the server response problem caused by forced recycling process when the access volume is high, resulting in IIS feigning dead and not responding)
2. Performance
only select idle timeout for 20 minutes. None of the others. The maximum number of working processes in Web park is 1 (default). Note that the default must be kept here in the web garden. If you fill in other numbers more than 1, the background program of some website programs will be unable to open or refresh continuously
the original request queue was limited to 4000, but now there is no limit< Third, running status
the first two items are used, which are the original default settings. The starting time is limited to 90 seconds and the closing time is limited to 180 seconds
remove the hook to start fast fail protection
in order to avoid that there is no prompt when many errors are encountered, you can not close it, but just increase the protection scope of quick protection. For example, if the number of failures is 50 and the time period is 5 minutes, close the corresponding program
"the closing time limit of 180 seconds" is necessary, because the closing time of the process was originally limited to 90 seconds, which is the default value. If the closing time of the process exceeds 90 seconds, it will be considered as a timeout, thus the process closing time exceeds the limit log, so it is appropriate to extend this time, This error can be avoided www.cnzz.cc
the second method:
reason: the memory heap of the independent process is consumed, IIS can't create more process workspace to handle
solution:
1. HKEY_ LOCAL_ MACHINE\ System\ CurrentControlSet\ Services\ W3SVC
2. Create a new DWORD under the parameters key, with the name: usesharedwpdesktop value of 1. Restart IIS
the third problem has been solved. It is found that the database connection cannot be released. I don't know what the reason is. The same code is good locally, and there is a problem at the server side. Finally, add the following statement in the connection string to solve the problem.
pooling = true; MAX Pool Size=512; Min Pool Size=50; Connection lifetime = 30
the fourth
creates a new application pool, and different websites reference different application pools
is an application pool; AppPool #1' The shutdown time of the serving process exceeded the limit. The process ID is; 3000'
after the above situation occurs, the website corresponding to the application pool is very slow to access and can hardly be opened
is this a feign death
restart the site, and the problem can be solved
maybe it's the application pool setting problem, I don't know how to set it
please give me some advice<
program code solution:
1 set the process pool recovery time. In the process pool attribute.
2 if your program uses ASP + ACC database, and the ACC database is larger than 30m, it is recommended to replace the SQL database. If the ACC database is larger than this value, it will cause frequent feign death of IIS6 in 2003 system, and it will cause dllhost.exe to occupy a lot of CPU and memory resources in 2000 system It will affect the speed of web access.
3 ASP program has a dead loop.
4 you can use Microsoft Office compression to repair the ACC database, and you must backup it first.
reference content 2003 application pool feign death common problems and Solutions
2006-10-09 09 09:48
we often see that 2003 application pool automatically dies and cannot be recovered, Service unavailable always appears. The common methods are as follows
1: before, this IIS6 would appear when there was no SP1 patch. 0 feign death problem, but now Microsoft are automatically updated inside the patch, generally you play the latest patch is not this problem Therefore, IIS feign death has little to do with this)
2: since IIS 6.0, CPU resources have been limited in the application pool, unlike previous IIS. 5 So the reason for the feign death pool is that the pool is pulled to death. When you can't open the website, you can see that one of your application pools is disabled, with a red cross on it. If you start the website with the right mouse button, it will recover automatically. This reason: it is probably caused by the following factors
(1): you limit the resource limit of the application pool too small, such as 50 or less. At this time, if the website under the pool takes up too much CPU, such as more than 50%, it will automatically die in 5 minutes. The application pool created manually by default does not operate when it exceeds the resources
the solution to the above situation: 1. Do not limit CPU resources. (this is not desirable, do not limit resources, some programs have bugs that take up a lot of resources, the server will be pulled dead, you may not be able to operate the server.) 2: Select close when the resource is exceeded. By default, this close will fail 5 times and recover within 90 seconds. Generally, it can be closed by default. The website can recover automatically. This shutdown: it is not a permanent shutdown. It means that the resource will be shut down and the pool will be automatically recovered within a certain period of time. No operation means no recovery. This is a misunderstanding of many people. Last time I wrote the related explanation
(2): memory is limited to IIS6. There are virtual memory and maximum memory limits on the 0 application pool, if you set this. So the site visits will also appear feign death, so it is not recommended to set here. By default
3: the memory of the server itself is too small. Of course, the website needs to use the memory. When the memory is not enough, the application pool will die and become disabled. Then the application pool can only be restored after all the memory is released. When this happens: then you have to consider adding memory or check what program is occupying memory. For example, the MSSQL database is a big memory user. It's better not to be on the same server as the web server at the same time. Many people use 1G memory to make 2003 system. The 2003 net structure takes up a lot of memory, so it's better to add 2G or higher memory to make 2003 server. Memory is not enough, the above two points mentioned, there is no way to operate, and can not automatically recover
4: even if the access database is too large or there are too many queries, IIS will be pulled to death. The solution is as follows:; Repair access database, or use access database as little as possible
5: different websites use different application pools: according to your own actual situation, it's better to have an independent application pool for a large site, which limits its resources beyond automatic recycling. See (1) above, so it doesn't affect other sites. Medium sized sites: multiple sites share an application pool, for example, five sites use one pool, set their resource time, and so on. In this way, even if they have super resources, they will not affect the websites of other application pools
6: set recycle time: many people think that the shorter the recycle pool is, the better. In fact, it is wrong. Each recycle of course reclaims the memory, but it increases the burden of the server once. When the server is busy, it may cause other application pools to die. So it is recommended to set a total of 1000. Other independent pools can be set to 600 according to the traffic of his website. It is not recommended to set the shared pool too short.
the server often generates "application pool"; DefaultAppPool' The shutdown time of the serving process exceeded the limit. The process ID is; 2068'” Is in the state of feign death. It is understood that the problem is the setting of IIS application pool. The solution is as follows:
Internet information services (IIS) manager > application pool > defaultapppool > right click Properties
1. Recycle work process (minutes): selected, value 1740
2. Recycle work process (number of requests): not selected (originally set to 35000)
3. Recycle work process at the following time: not filled
4 Recycle worker processes when consuming too much memory: none 2. 3 and 4 items may avoid the server response problem caused by forced recycling process when the access volume is high, resulting in IIS feigning dead and not responding)
2. Performance
only select idle timeout for 20 minutes. None of the others. The maximum number of working processes in Web park is 1 (default). Note that the default must be kept here in the web garden. If you fill in other numbers more than 1, the background program of some website programs will be unable to open or refresh continuously
the original request queue was limited to 4000, but now there is no limit< Third, running status
the first two items are used, which are the original default settings. The starting time is limited to 90 seconds and the closing time is limited to 180 seconds
remove the hook to start fast fail protection
in order to avoid that there is no prompt when many errors are encountered, you can not close it, but just increase the protection scope of quick protection. For example, if the number of failures is 50 and the time period is 5 minutes, close the corresponding program
"the closing time limit of 180 seconds" is necessary, because the closing time of the process was originally limited to 90 seconds, which is the default value. If the closing time of the process exceeds 90 seconds, it will be considered as a timeout, thus the process closing time exceeds the limit log, so it is appropriate to extend this time, This error can be avoided www.cnzz.cc
the second method:
reason: the memory heap of the independent process is consumed, IIS can't create more process workspace to handle
solution:
1. HKEY_ LOCAL_ MACHINE\ System\ CurrentControlSet\ Services\ W3SVC
2. Create a new DWORD under the parameters key, with the name: usesharedwpdesktop value of 1. Restart IIS
the third problem has been solved. It is found that the database connection cannot be released. I don't know what the reason is. The same code is good locally, and there is a problem at the server side. Finally, add the following statement in the connection string to solve the problem.
pooling = true; MAX Pool Size=512; Min Pool Size=50; Connection lifetime = 30
the fourth
creates a new application pool, and different websites reference different application pools
is an application pool; AppPool #1' The shutdown time of the serving process exceeded the limit. The process ID is; 3000'
after the above situation occurs, the website corresponding to the application pool is very slow to access and can hardly be opened
is this a feign death
restart the site, and the problem can be solved
maybe it's the application pool setting problem, I don't know how to set it
please give me some advice<
program code solution:
1 set the process pool recovery time. In the process pool attribute.
2 if your program uses ASP + ACC database, and the ACC database is larger than 30m, it is recommended to replace the SQL database. If the ACC database is larger than this value, it will cause frequent feign death of IIS6 in 2003 system, and it will cause dllhost.exe to occupy a lot of CPU and memory resources in 2000 system It will affect the speed of web access.
3 ASP program has a dead loop.
4 you can use Microsoft Office compression to repair the ACC database, and you must backup it first.
reference content 2003 application pool feign death common problems and Solutions
2006-10-09 09 09:48
we often see that 2003 application pool automatically dies and cannot be recovered, Service unavailable always appears. The common methods are as follows
1: before, this IIS6 would appear when there was no SP1 patch. 0 feign death problem, but now Microsoft are automatically updated inside the patch, generally you play the latest patch is not this problem Therefore, IIS feign death has little to do with this)
2: since IIS 6.0, CPU resources have been limited in the application pool, unlike previous IIS. 5 So the reason for the feign death pool is that the pool is pulled to death. When you can't open the website, you can see that one of your application pools is disabled, with a red cross on it. If you start the website with the right mouse button, it will recover automatically. This reason: it is probably caused by the following factors
(1): you limit the resource limit of the application pool too small, such as 50 or less. At this time, if the website under the pool takes up too much CPU, such as more than 50%, it will automatically die in 5 minutes. The application pool created manually by default does not operate when it exceeds the resources
the solution to the above situation: 1. Do not limit CPU resources. (this is not desirable, do not limit resources, some programs have bugs that take up a lot of resources, the server will be pulled dead, you may not be able to operate the server.) 2: Select close when the resource is exceeded. By default, this close will fail 5 times and recover within 90 seconds. Generally, it can be closed by default. The website can recover automatically. This shutdown: it is not a permanent shutdown. It means that the resource will be shut down and the pool will be automatically recovered within a certain period of time. No operation means no recovery. This is a misunderstanding of many people. Last time I wrote the related explanation
(2): memory is limited to IIS6. There are virtual memory and maximum memory limits on the 0 application pool, if you set this. So the site visits will also appear feign death, so it is not recommended to set here. By default
3: the memory of the server itself is too small. Of course, the website needs to use the memory. When the memory is not enough, the application pool will die and become disabled. Then the application pool can only be restored after all the memory is released. When this happens: then you have to consider adding memory or check what program is occupying memory. For example, the MSSQL database is a big memory user. It's better not to be on the same server as the web server at the same time. Many people use 1G memory to make 2003 system. The 2003 net structure takes up a lot of memory, so it's better to add 2G or higher memory to make 2003 server. Memory is not enough, the above two points mentioned, there is no way to operate, and can not automatically recover
4: even if the access database is too large or there are too many queries, IIS will be pulled to death. The solution is as follows:; Repair access database, or use access database as little as possible
5: different websites use different application pools: according to your own actual situation, it's better to have an independent application pool for a large site, which limits its resources beyond automatic recycling. See (1) above, so it doesn't affect other sites. Medium sized sites: multiple sites share an application pool, for example, five sites use one pool, set their resource time, and so on. In this way, even if they have super resources, they will not affect the websites of other application pools
6: set recycle time: many people think that the shorter the recycle pool is, the better. In fact, it is wrong. Each recycle of course reclaims the memory, but it increases the burden of the server once. When the server is busy, it may cause other application pools to die. So it is recommended to set a total of 1000. Other independent pools can be set to 600 according to the traffic of his website. It is not recommended to set the shared pool too short.
3. Are you sure it's the win7 driver, not the win10 driver? You can delete it if you are sure. It's better to stay, if it doesn't take up a lot of space
4. It should be that some IIS service of the system has been opened.
5. You can set the recycle time if there is a problem with the program. It can be set by intensive program recycling time<
how to restart after stopping
there is a resource limitation operation
which exceeds
1: no operation
2: close
this non operation and close is misunderstood by many people. You may think that if you don't operate it, you will achieve this resource. If I don't limit it, it will be OK and won't die. But you are wrong. This non operation means that the pool will be dead automatically when the resource is reached, and the recovery will not be restarted automatically when the resource is dead. So we have to close it here. After closing the default is 90 seconds after automatic recovery
one program pool per site??? If there are many sites, you certainly don't have enough resources,
suggest 4-5 sites have one pool.
how to restart after stopping
there is a resource limitation operation
which exceeds
1: no operation
2: close
this non operation and close is misunderstood by many people. You may think that if you don't operate it, you will achieve this resource. If I don't limit it, it will be OK and won't die. But you are wrong. This non operation means that the pool will be dead automatically when the resource is reached, and the recovery will not be restarted automatically when the resource is dead. So we have to close it here. After closing the default is 90 seconds after automatic recovery
one program pool per site??? If there are many sites, you certainly don't have enough resources,
suggest 4-5 sites have one pool.
6.
-
control panel - programs and functions - turn on or off windows functions, Internet information services, web management tools, and all the files in Microsoft. Net framework, and then wait a few minutes for windows configuration to finish, Then it's OK
< / OL >
remarks: if classic. Net apppool can't be used after adding it, just check the Internet information service in the first step and the files in Microsoft. Net framework. I don't know why. Anyway, it's ok after that. Good luck
7. W3wp.exe is a Windows system file. The program has no visual window. This is a trusted program released by Microsoft
8. It is an application pool; DefaultAppPool' The process providing the service terminated unexpectedly. The process ID is; 3056' Process exit code is & # 39; 0xffffffff'
for more information, see the http://go.microsoft.com/fwlink/events.asp Our help and support center<
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -,
select "IIS admin service" under it, right-click to select "properties", find "security", edit "custom" in "startup and activation permissions", add the account "network service",
give the account "local startup" and "local activation" permissions, restart IIS (click "start" - "run" - "CMD", click OK, and then click OK, Then run iisreset),
and then visit the same site, everything is normal<
another solution:
event type: warning
Event Source: W3SVC
event type: none
event ID: 1009
date: 2007-8-xx
Event: XX: XX
User: XX
Computer: XXX
Description:
is an application pool; XXXXX' The process providing the service terminated unexpectedly. The process ID is; XXXX' Process exit code is & # 39; 0x80'
for more information, see the http://go.microsoft.com/fwlink/events.asp Our help and support center<
cause:
together with each worker process that IIS creates under a separate identity, the system creates a new desktop object by allocating memory from the configured desktop heap. This issue occurs because, when that heap has been exhausted, IIS cannot create more worker processes. Clients then receive the " service unavailable" Error message in their web browsers when they try to visit web sites that those application pools host.
the memory heap of the independent process is consumed, IIS cannot create more process workspace to handle
solution:
warning: you need to modify the server registry, This registry key permits all worker processes to run in one shared desktop, registrarless of their worker process identities.
to add the usesharedwpdesktop registry key:
1. HKEY_ LOCAL_ MACHINE\ System\ CurrentControlSet\ Services\ W3SVC
2. Create a new DWORD key under the parameters key. The name is: usesharedwpdesktop value is 1. Restart IIS
MS about this key value Description:
usesharedwpdesktop
registry path: HKEY_ LOCAL_ MACHINE\ System\ CurrentControlSet\ Services\ W3SVC\ Parameters
data type: reg_ DWORD
Default: 0
range: 0 - 1
if you are setting up application pools with a unique ID, the maximum number of application pools will be about 60, depending on the application and memory resources on the server. Some system resources with a single new login session are limited. This means that there can be 60 processes running at the same time under different accounts. IIS 6.0 supports running these processes in a single shared workstation and desktop at the cost of sharing a single encapsulation of a single user session between all parties. To scale to more than 60 application pools and share a single desktop, change the useshared wpdesktop to a DWORD value of 1. After changing this registry key, it should be able to expand to hundreds of application pools and hundreds of working processes running at the same time
the above is a reprint, whether it returns to normal still needs to be verified
continued: after the above two methods, the system log is still in error, which means that the process related to the application pool (communicating with the World Wide Web) is a fatal error. Suddenly, I saw a sentence on the Internet, saying that I want to give the user permission of system32 (I guess I want to give system32 & # 92; Temps, because it's the NT version of discuz). After giving it, the website can do it, but the error will still appear. Finally, according to the first method mentioned above, we found that the NetMan's network service permission was missing. After adding it, after several hours of observation, the error did not come out
for more information, see the http://go.microsoft.com/fwlink/events.asp Our help and support center<
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -,
select "IIS admin service" under it, right-click to select "properties", find "security", edit "custom" in "startup and activation permissions", add the account "network service",
give the account "local startup" and "local activation" permissions, restart IIS (click "start" - "run" - "CMD", click OK, and then click OK, Then run iisreset),
and then visit the same site, everything is normal<
another solution:
event type: warning
Event Source: W3SVC
event type: none
event ID: 1009
date: 2007-8-xx
Event: XX: XX
User: XX
Computer: XXX
Description:
is an application pool; XXXXX' The process providing the service terminated unexpectedly. The process ID is; XXXX' Process exit code is & # 39; 0x80'
for more information, see the http://go.microsoft.com/fwlink/events.asp Our help and support center<
cause:
together with each worker process that IIS creates under a separate identity, the system creates a new desktop object by allocating memory from the configured desktop heap. This issue occurs because, when that heap has been exhausted, IIS cannot create more worker processes. Clients then receive the " service unavailable" Error message in their web browsers when they try to visit web sites that those application pools host.
the memory heap of the independent process is consumed, IIS cannot create more process workspace to handle
solution:
warning: you need to modify the server registry, This registry key permits all worker processes to run in one shared desktop, registrarless of their worker process identities.
to add the usesharedwpdesktop registry key:
1. HKEY_ LOCAL_ MACHINE\ System\ CurrentControlSet\ Services\ W3SVC
2. Create a new DWORD key under the parameters key. The name is: usesharedwpdesktop value is 1. Restart IIS
MS about this key value Description:
usesharedwpdesktop
registry path: HKEY_ LOCAL_ MACHINE\ System\ CurrentControlSet\ Services\ W3SVC\ Parameters
data type: reg_ DWORD
Default: 0
range: 0 - 1
if you are setting up application pools with a unique ID, the maximum number of application pools will be about 60, depending on the application and memory resources on the server. Some system resources with a single new login session are limited. This means that there can be 60 processes running at the same time under different accounts. IIS 6.0 supports running these processes in a single shared workstation and desktop at the cost of sharing a single encapsulation of a single user session between all parties. To scale to more than 60 application pools and share a single desktop, change the useshared wpdesktop to a DWORD value of 1. After changing this registry key, it should be able to expand to hundreds of application pools and hundreds of working processes running at the same time
the above is a reprint, whether it returns to normal still needs to be verified
continued: after the above two methods, the system log is still in error, which means that the process related to the application pool (communicating with the World Wide Web) is a fatal error. Suddenly, I saw a sentence on the Internet, saying that I want to give the user permission of system32 (I guess I want to give system32 & # 92; Temps, because it's the NT version of discuz). After giving it, the website can do it, but the error will still appear. Finally, according to the first method mentioned above, we found that the NetMan's network service permission was missing. After adding it, after several hours of observation, the error did not come out
9.
This is the application pool,
the application pool is the configuration that links one or more applications to one or more worker process collections
can be configured in the window function
10. If you have time to continue, you just want to post
Hot content