Hi colleagues, My company is recently struggling with 1C enterprise performance. It takes 1-3min after running client (thick,thin,web) to get the window with users list and after typing user it takes another 1-2min to login to application.
We have databases hosted on SQL server. Database and terminal servers (two) are within company domain. Web server and application servers (two cluster servers) are in DMZ domain. We have a separate license server (company domain).
It is hardly to find out if slow startup is cause by delayed response (from DMZ we cannot ping license server or SQL server). Hardware usage on all servers is normal, SQL DBs have maintenance plan running daily (reindexing,update of statistics).
Can we somehow based on logs or by running developer mode check what is causing such slow response while logging?
After we login to 1C, performance is quite good. So it's mostly affecting starting the clients and logging.
I think you have some problems with the availability of a domain controller from the 1C server. So 1C service tries to use the operation system authentication and hangs for a while.
Hi, Sorry for being late in my response. We have tried to run some of infobases with no windows authentication parameter, but there was no improvement. It took exactly the same time to start the application (~3-4min).
Can we somehow find some hints in the logs which may lead us to the answer of that issue? I have suspicion that slow startup is related to the license server and it's response..
Hi Michal, There are a couple of useful links - unfortunately, in Russian, and unfortunately, with some admin skills needed. <deleted as rules violation>
To make long story short, in both cases guys switched off IPv6 at their servers, and it helped.
Could you, please, answer a few questions to help me understand the issue?
1. What version of the Platform do you use? 2. Was it always that slow or the issue's arisen at some point? 3. If you run the second 1C client when the first one has already started will it take the same time? 4. Does 1C client start on the local computer or on the terminal server you've mentioned? 5. Are 1C and SQL servers virtual or real?
And the most important thing. I need you to collect the tech log that will hopefully show us the problem. These are the steps to follow: 1. Copy the attached logcfg.file into the "Program Files (x86)\1cv8\conf" folder next to the "Program Files (x86)\1cv8\bin" folder 1C client starts from. Please, note that the file needs to be on the computer you physically run 1C client on. 2. Run 1C client, wait until it's started. 3. Archive the content of "c:\1c\logs" folder and post it here for me.
Hi All, I am sorry for being late in my response. I have attached picture of our infrastructure for 1C servers. License is acquired from another server (hardware license). Please see my comments below.
@Eugeny Mamontov "To make long story short, in both cases guys switched off IPv6 at their servers, and it helped." IPv6 is already switched off - no results.
@Konstantin Rupasov 1. What version of the Platform do you use? 8.3.8.1964 2. Was it always that slow or the issue's arisen at some point? It was not slow at the beggining (that time we had ~5-8 databases). 3. If you run the second 1C client when the first one has already started will it take the same time? Yes it takes the same time. 4. Does 1C client start on the local computer or on the terminal server you've mentioned? Slow startup occurs on all servers. It's same time to start it on APP server, terminal or Web (slightly faster). 5. Are 1C and SQL servers virtual or real? 1C and SQL are virtual machines.
CONN.zip contains application logs. It's in use almost all the time so it was hard to generate logs just for my connection (I tried to filter them to last few minutes when I start the application)
Hi Michal, There are a couple of useful links - unfortunately, in Russian, and unfortunately, with some admin skills needed. <deleted as rules violation>