Gemini running slow/locking up since...
Gemini 3.6 has started to run really slow or locking up since adding in the Outlook Add-On on one machine. I've uninstalled the outlook add-on but Gemini still seems "hosed".
englea
· 1 |
|
Tuesday, January 19, 2010, 3:58:51 PM |
0
|
Is that the web application that is running slow? |
||||
|
0
|
yes, the web app. it's intermittent. really, what it's doing is "hanging" and becoming non-responsive. but not all the time. it's hard to explain, but it JUST started doing it right after the Outlook Add-On install/synchronization. Database size is 23 MB. Hope that's not a problem... :) |
||||
|
0
|
It might be that the data is fragmented. Try and update statistics on the geminiissues table: |
||||
|
0
|
completed. for now it seems to be working better, but like I mentioned it's intermittent. will keep posted. great system v3.6. |
||||
|
0
|
it hangs when entering a new task, regardless of project it's being entered under. i double-checked to see if closing the time tracker app or login out of the vs plug-in helped, neither did. |
||||
|
0
|
Which version of SQL server are you using? |
||||
|
0
|
SQL 2005 SP3 We cannot enter new tasks. The application hangs there until it throws a user/password challenge. The task still goes in the database (if we let it hang long enough), but none of the users can enter a new task. Like I said, this started yesterday when I installed the Outlook Add-On and did a synchronize. That's when it started hanging. Before that it was fine. |
||||
|
0
|
Can you please try and restart the web app? We haven't seen this issue before. Did you try another browser? |
||||
|
0
|
i thought I mentioned that, but i guess i didn't. I've restarted the app (three times), the app pool (several), and the server (once). another browser isn't an option at our location (policy). no other software was installed after the outlook add-on. and this started RIGHT after I installed that add-on after the first calendar synch. I can't narrow it down to much more than that... |
||||
|
0
|
we re-indexed the entire database and that seems to have fixed the issue. i started noticing that i could add a new task to some projects, but not others. i'm able to add tasks to the ones I wasn't able to before the re-indexing. if anything changes...i'll post it here. thx! |
||||
|
0
|
Please check the database increment settings and make sure its not too small. |
||||
|
0
|
what's considered too low for a 22mb database? the autogrowth for the db is 10mb, the log is 10%. default settings. still kinda having problems: i can delete a task from a list of tasks/issues, but when I click to delete a second one, it hangs. I noticed this deleting all of the "test" tasks I created.
EDIT: you might want to explain exactly what you want us to check because the dba is a bit confused if you're talking about "column increment settings" or the auto-growth, or whatever else you may be talking about. |
||||
|
0
|
The autogrowth is too low. 10% of the initial size was a tiny amount and thus the database got very fragmented. |
||||
|
0
|
backed up, restored, we've re-indexed the entire db (as noted earlier)...still hangs. we're on deadline now...so the priority just jumped to get this fixed. i know you've never "had this issue before", but I can't stop thinking about how it was running like a dream (best ever) until i installed the Outlook Add-On and clicked "synchronize", then it all went south, and has been south ever since. It's usable, only until it decides to "hang" and it loses connection (apparently). We're no longer thinking it's a db issue. Is there something we need to check in the app, web.config, somewhere? We're not having any issues with any of the other apps running on the same server, fyi, so that rules out network, hardware, server, and IIS 7. And it occurs regardless of user/machine. Just Gemini, with the issue...since the outlook add-on & synch. |
||||
|
0
|
Please send an email to support at countersoft dot com. We will remote in to your machine to see if we can fix this. |
||||
|
0
|
I'm not trying to be complicated, but remoting in isn't an option due to the nature of our company business (confidentiality, etc). Yes, as stated (above), I have uninstalled the outlook add-on. however, the vs2008 plug-in and time tracker apps are running, currently. but i did some tests with both of those closed and got the same results. it hangs. AND, it seems to hang on the newly created projects...which I thought was odd. I did create several new projects on Monday (but hadn't had this problem and/or hadn't noticed it because no work was being done on those projects, etc). NOTE: I just did a test on a project that has been in Gemini for a year, created a new task, and it hung. That was as far as I was able to get. |
||||
|
0
|
We need to establish why it hangs and on what exactly. Is it on the database connection, execution or the web request. |
||||
|
0
|
still hanging. intermittent. Just tried to add new task, fail. the task is in the system but the application didn't move forward. server event logs show processes "quitting" (worker processes from application pool). Nothing showed abnormal on a profiler. Application is pretty much useless at this point and things are really piling up (sticky notes to be exact).
I am going to take it to 3.6.3 and cross fingers. |
||||
|
0
|
upgraded to 3.6.3 went into a task, logged time (1 entry), marked as completed, marked as closed and it hung (didn't mark as closed). after about 30 seconds it returned to the home page (not the project home page, but the root page). From the application log: Faulting application w3wp.exe, version 7.0.6002.18005, time stamp 0x49e023cf, faulting module ntdll.dll, version 6.0.6002.18005, time stamp 0x49e03821, exception code 0xc0000374, fault offset 0x000afaf8, process id 0x1298, application start time 0x01cab19c75e351e3.
From System Log: Event 5010, WAS A process serving application pool 'gemini' failed to respond to a ping. The process id was '4760'. This is on Server 2008, SQL 2005, IIS 7 (as mentioned). What should the application pool look like for IIS 7? Any special mods? |
||||
|
0
|
SQL Profiler is rather interesting: My database guy has Gemini home page open...not clicking on anything...and the profiler is generating hits (many hits) to the database with no activity on the website at all. Is this normal? |
||||
|
0
|
The application pool should be integrated (not classic). It is not normal for the Gemini database to get many hits if you are sure no one else was using it. What are the SQL commands? Do you have the scheduler running? |
||||
|
0
|
Not sure what you mean by "What are the SQL commands". I don't think we set up scheduler. How can I tell? None of your add-on apps are installed or running. SQL profiler reported this when all we had open was the main page for a project: 12,000 hits in 12 minutes to Gemini database. Is this excessive? |
||||
|
0
|
You shoud be able to see what are the SQL commands in the profiler. 12k hits for an idle app is way too much, there must be some other prcoess that is accessing the database.Are you sure that all these hits are from Gemini |
||||
|
0
|
all Gemini. We do not get that from any of the other apps, and it's running on our dev server so the only ones that hit that SQL server are us (3 people). We might have clicked into one of the projects and back out. But we weren't in there creating tasks and what not. What other thing could be running? Some leftover code from one of your add-on apps? Especially the Outlook add-on? (that's where this whole thing started). |
||||
|
0
|
As a test to see if it is the add-ons, remove / rename the API folder in the Gemini virtual directory. If you are still getting heavy load then it is not that. |
||||
|
0
|
thought you might have had it with this one, but nope. I've deleted the IIS 7 website and app pool, and recreated both. let me run with that to make sure it's not IIS7's fault. |
||||
|
0
|
Gemini started running SUPER fast with out fail. I was able to create a task, log in two time evnets, mark it "closed", and bam, as soon as I tried to mark it "complete" it locked. System Event Log shows: WARNING: A process serving application pool 'gemini' failed to respond to a ping. The process id was '7992'. WARNING: A process serving application pool 'Gemini' suffered a fatal communication error with the Windows Process Activation Service. The process id was '2844'. The data field contains the error number. double-checked (and fixed) a pile of things in IIS 7 (like the apppoolidentity on some pools). had it running like a dream again for a couple of minutes...locked/hanged on create issue. so...i don't know now. |
||||
|
0
|
These posts might help with debugging this: http://technet.microsoft.com/en-us/library/cc735271(WS.10).aspx Did you check Gemini's error log table for errors? Also, is it a specific action that causes this? |
||||
|