Restarting IIS/WWW Service Daily after 9.5

Do you have a support ticket open on this issue?

No. I'm contemplating reinstalling my Orion completely.

smcdonald

I have seen the same problem - I see this randomly -

This seems to happen when I have a lot of windows open like 5-10 and even then not always. The only fix I have found is the same as yours - restarting IIS. I also noticed the WWW publishing service using up lots of memory - mine goes up as high as 320MB before I tend to see the above error. Already got a ticket in for this since last week...

Server Error in '/' Application.

Timeout expired. The timeout period elapsed prior to obtaining a connection from the pool. This may have occurred because all pooled connections were in use and max pool size was reached.

No. I'm contemplating reinstalling my Orion completely.

If you have time, please open a Support ticket and include Diagnostics. That's the only way we're likely to find the root cause.

I just opened a case for this. Thanks. Any ideas?

brahuna

I am also having this issue. The webpage is blank and just keeps trying to load. I installed SP1 this morning, and it stopped again this afternoon. Had to restart the WWW service.

clanger

We also have this error - same web error as someone else posted.

Interestingly, I'd posted a link to my boss to a particular interface on a switch. Even with the main page (or any other page I had open) displaying the error text someone posted earlier, he was able to get to the switch interface page from the direct link I'd sent him and it still worked.

But yes, we are restarting daily (at least) as well since 9.5 upgrade (also applied 9.5 sp 1)
I WAS restarting everything, but I'll try just the web from now on to resolve it.

I'm not getting an update from support. Can anyone from SolarWinds provide any information?

chris.schear

We aren't running 9.5, we're running 9.1 SP5, but we have to go to the Orion Service Manager and Unload COM+ Application every morning. On some days, we have to do it multiple times during the day, as well.

ElisabethZakes

DirtySouth, the case has been escalated to Development, and they're working on a fix. Your case support tech will certainly let you know when there are updates.

KCWayner

I'm in the same boat. I opened a case last week and have sent a few updates with diagnostics but haven't heard anything in the last few days. Case 99007. There were some errors in the logs that I included too not knowing if they had anything to do with it.

ElisabethZakes

KC, I'll have a support tech look at your case.

brahuna

Has there been any resolution on this? We are still having to restart the WWW service at least once per day.

It's scheduled for an upcoming SP.

lchance

I only starting having this problem the past two days, long after my NPM 9.5 update and the SP1.

I've been restarting everything in Orion System Manager to clear it up. I also opened a case on this earlier today and sent in a diagnostic file.

morgando

We too are having this issue. Curiously one of my guys here has been using Firefox. I have since loaded it and it seems to weather these hangs far better than IE (6.0.2900 corp version)

When we do experience these issues it appears that all navigation from the Network Summary home page "hangs". SP3 anytime soon??

I too have experienced this and other issues with the v9.5 upgrade.

I opened a case on this specific issue on 6/28/2009, case #103164. I was told by support that no one else was having this issue and that Orion was working fine - it must be my webserver. With v9.5 SP1 running my webpage will stop loading about every 36 hours - fix is as you have described. When I apply SP2 my webpage will stop loading about every 20 minutes. I can go back to base 9.5 and duplicate this behavior at will.

I have performed numerous troubleshooting techniques with my support rep and have more outstanding they want me to do - have never been told this is a known issue LET ALONE that it is being worked on for SP3. I just emailed this thread my my technician. Anxiously awaiting the fix as this has virtually crippled our use of NPM.

Thanks for your initial note.

I just talked to support and they did acknowledge the issue. I am also anxiously awaiting a fix; we have problems at least once a day.

morgando

Has anyone had this issue where Firefox is being used? We had to reset three times last weekend. A co-worker using FF was not having the nagging hangs. I have since loaded FF 3.5 and NO issues since Sunday.... and i have been abusing it.....

jcooler

Has anyone had this issue where Firefox is being used? We had to reset three times last weekend. A co-worker using FF was not having the nagging hangs. I have since loaded FF 3.5 and NO issues since Sunday.... and i have been abusing it.....

I use Firefox almost exclusively and have experienced this problem quite a bit. When the site is down, its down for users on any browser until IIS is restarted.

I've been able to manually recycle the worker process by ending the w3wp.exe process with task manager. It's slightly quicker than starting and stopping IIS completely.

We have roughly 10 users logged in at any given time during the day. We are all using IE and when the webpage poops, it affects everyone at the same time. I'm not sure how that could be browser related.

AlbanyNYMike

I agree withthe consensus that it should not matter what browser is being used, as onec the website goes, it goes for everyone.

This does seem to be directly related to the w3wp.exe as everytime the website craps out, w3wp is running around 200 - 300,000k in the task manager.

Ending w3wp seems to do the trick, but to be cautious, I open Orion Service manager and stop / start the IIS website NPM and NCM.

Anyone try just stopping w3wp and not going into the OrionServiceManager?

Anyone having SWJobEngineWorker.exe running over 250,000k?

SP3 where oh where are you??

I have not completely reinstalled yet but I may very well try that.

I have repaired base 9.5 and installed SP1 - I could go about 1-1.25 days and the problem would reoccur.

Before loading the hotfix I am experimenting once more - this morning I loaded SP2 to make sure I was not crazy and so far my website has stopped 2 times within an hour each time - the W3WP process starts around 115000K and quickly (within 10 minutes) jumps to around 175,000K ultimately getting over 200,000K.

For grins I ended W3WP in task manager and it autostarted a new process within seconds and things are good again.

Once the site fails again I will load the hotfix and post an update.

I have not completely reinstalled yet but I may very well try that.
I have repaired base 9.5 and installed SP1 - I could go about 1-1.25 days and the problem would reoccur.
Before loading the hotfix I am experimenting once more - this morning I loaded SP2 to make sure I was not crazy and so far my website has stopped 2 times within an hour each time - the W3WP process starts around 115000K and quickly (within 10 minutes) jumps to around 175,000K ultimately getting over 200,000K.
For grins I ended W3WP in task manager and it autostarted a new process within seconds and things are good again.
Once the site fails again I will load the hotfix and post an update.

How many users do you typically have logged into Orion when your's stops working? I've noticed that ours runs without issue in the evenings and overnight. At that time we only have 1-2 users logged in at any given time.

Before reinstalling, please apply SP2 and apply the hotfix and see if that resolves your issue.

I can have this issue with only 1-2 folks logged in. From watching the memory usage it seems the first 1-5 page refreshes cause the biggest jump in memory usage. Once past these first few it seems to even out but then will ultimately stop.

SP2 and hotfix loaded - initial memory jumps observed same as before.

W3WP hovering around 171,176K

Working ok so far - 15 minutes and counting...

Rojon

We just tried applying the Hotfix and now we're getting this Plugin Error: "An error occurred inside a plug-in contained on this page." each time we log into the web console.

Has anyone had success running the Hotfix???

Rojon

Rojon,

Can you please open a support ticket and reference this thread and upload diagnostics so we can look into further.

fast

I'm having the same problem after upgrading to 9.5 although I need to restart WWWPS multiple times a day. I applied the hotfix Friday and that made no difference.

I avoided the upgrade for as long as possible but a license upgrade (SL2000 to SLX) forced me into this. Too bad because it's been a fiasco ever since.

Can we get some commentary on why this hotfix isn't making a difference for a lot of people in this thread?

Update - On Saturday I went ahead and removed all maps (2 of them) on my homepage. I reran the config wizard effectively uninstalling the hotfix - only SP2 now.

My system has been running fine since - certainly would indicate a problem with map presentation...

Of course I want my maps displayed so fixing this is certainly still an issue.

I am scheduled to connect with tech support today to further discuss.

We are looking into, this is why I have requested diagnostics and a support case, it is too difficult to speculate without more data

KCWayner

I'm still on SP1 as I wanted to see how SP2 worked for people. My site was crashing too with just SP1 (ticket 99007). I then created a batch command to stop the w3wp process every day at 1 AM. Have not had the web page lock up on me since. It has been a week since I created it. The memory usage still goes high, but I guess it stays down enough to work. We only have 2 users in on a regular basis. Granted, this is not a fix, it is a work-around. Still waiting before I consider going to SP2 and whatever else is needed.

timf

thought i posted already but i don't see it out there.

I have found that I don't need to do a full service restart to get my site working again. I can go into IIS manager, expand application pools folder, right-click on defaultAppPool and hit recycle. It's instant and more often than not it will bring my website back up.

Jim_McCormick

Yesterday, I removed our one and only map from the Network Summary Page. I have not had to restart the WWW service since. Since our upgrade to 9.5 SP2, this is the longest we've gone without having to restart the service. If this uninterrupted service continues, there would be no doubt that the map is the cause of the problem.

KCWayner

I bit the bullet and went ahead with the SP2 and the hotfix and wow, my map loads 100x faster than it did before. I have kept my maps pretty simple and since the upgrades I have redone most of them to take advantage of options I hadn't used before. I have disabled my scripts that killed the w3wp process to see if that is needed anymore. Now I just need to figure out the reports that have broken due to the upgrades.

jlgugler

I have lots of problem since the 9.5 sp2

I have to restart the web server and/od the service

the polling process seem to "hang" sometime. i get some errors on the servers and some services stops.

Arte there already some patch ?

Please contact Support and open a case on this issue and they can supply you with the hotifx

morgando

We have removed the map from our home page and we have not had to restart services yet. I am noting huge memory utilization on the following

w3wp.exe = 753M (roughly 5500 elements on this polling engine)

SolarWinds.BusinessLayerHost.exe = 1.4G

Not seeing much over about 5% CPU on any one process at a time.....

Any word on SP or updated HF?

brian_duvall

I am having same issue with BusinessLayerHost.exe going up to 1.6G of memory util. Opened case 102994 and support provided me with "hotfix" to replace two map web files that cause memory leaks. Once the hotfix files were in place and IIS restarted my maps would cease to load and would get an error "System.ServiceModel.Security.SecuritySessionClientSettings`1+ClientSecurityDuplexSessionChannel[System.ServiceModel.Channels.IDuplexSessionChannel], cannot be used for communication because it is in the Faulted state."

Still working with support but had to remove "hotfix" since it caused more glaring problems than the memory leak of the Orion Service Module to my end users.

bump...

any updates from SolarWinds on SP3?

wwright00

No word yet. I wonder if there's a limitation in how many imbedded Maps you can have on your main view. i just built a brand new overview using Network Atlas that has 7 imbedded maps and i get this error when trying to display it on the main page:

Error: The communication object, System.ServiceModel.Security.SecuritySessionClientSettings`1+ClientSecurityDuplexSessionChannel[System.ServiceModel.Channels.IDuplexSessionChannel], cannot be used for communication because it is in the Faulted state.

i have another map that has 13 embedded maps and have been unable to get it to display. Has anyone opened a Solarwinds Case on this?