Separate names with a comma.
Discussion in 'General Discussion' started by aisagtr, Jan 19, 2008.
cpsrvd also checks the status of Apache by default, hence the entries in your Apache logs.
I'm actually talking about the Apache Status page, not the logs.
Does cpsrvd check the Apache status 8 times within a 8 second span?
See what I mean here? I understand what you are saying, but it doesn't appear these entries follow the patten that would be created by a regular cpsrvd check every so many minutes. If I saw these Apache requests spaced out 8 mins apart, then this conversation would be over. But that simply is not what is happening.
Here is a cut & paste from my status page. Look at the SS (Seconds since beginning of most recent request) column (these are the #'s in the 8000's):
17-5 - 0/0/103 . 0.00 8650 0 0.0 0.00 1.07 127.0.0.1 my.host.name OPTIONS * HTTP/1.0
18-5 - 0/0/560 . 0.00 8649 0 0.0 0.00 2.73 127.0.0.1 my.host.name OPTIONS * HTTP/1.0
19-5 - 0/0/647 . 0.00 8648 0 0.0 0.00 3.41 127.0.0.1 my.host.name OPTIONS * HTTP/1.0
20-5 - 0/0/331 . 0.00 8647 0 0.0 0.00 1.00 127.0.0.1 my.host.name OPTIONS * HTTP/1.0
21-5 - 0/0/1593 . 0.00 8644 0 0.0 0.00 12.06 127.0.0.1 my.host.name OPTIONS * HTTP/1.0
22-5 - 0/0/8 . 0.00 8643 0 0.0 0.00 0.00 127.0.0.1 my.host.name OPTIONS * HTTP/1.0
23-5 - 0/0/1325 . 0.00 8646 0 0.0 0.00 8.56 127.0.0.1 my.host.name OPTIONS * HTTP/1.0
24-5 - 0/0/1118 . 0.00 8645 0 0.0 0.00 8.91 127.0.0.1 my.host.name OPTIONS * HTTP/1.0
You can clearly see there was a request every second for an 8 second span.
Why would cpsrvd do that? And if it did, why wouldn't any of the slots be reused?
I fully understand that these are OLD entries, they are not currently active, they happened over 8000 seconds ago. I'm trying to understand why they exist.
I have this same problem but it only started once I used Easy::Apache v3.2.0 Build 3990 to remove my installation of TomCat.
Did any one else have TomCat installed and then removed it?
Maybe that has something to do with it, maybe not.
I never had TomCat installed but I have this issue.
If this issue is just cPanel checking to make sure Apache is still up then why would I have entries in my Apache status page showing requests spaced just 1 second apart for several seconds in a row? cPanel checks the status of Apache ever few minutes, not ever second, right?
At this point I do not believe these are being created by cPanel checking the status of Apache. But if that is the explaination, then something is wrong with cPanels' Apache checking routine.
I also had Apache 2.2 installed and once I downgraded to Apache 2.0 all of the HTTP/1.0 are gone.
Wondering if someone got any info from cpanel about this?
Mine is working out good with Apache 2.0 Although 2.2 would be much better to use.
I want to add something. I have looked over the forums, I contacted my support and I have search the net and everyone seems to say the same thing that having all of the HTTP 1.0showing up is not a big thing. Well I recently had an issue with the MySQL load being very high and I contacted my support and I also looked on the forums and both say the same thing, check your "Apache Status" to see what may be creating the draw.
Who else has received this kind of an answer?
I wish Cpanel would do something about this because who can check there Apache status when it is filled with HTTP 1.0? Doesn't anyone from the development team think about this? These 100+ HTTP fill up the available area and hurt server admins in dealing with there daily issues.
I'm not sure the slots with these requests in them are not reused by Apache. I'm not sure that is the case. But something is creating them... maybe some kind of lame attempted hack or DoS attack, not sure... but it doesn't seem to be what others have suggested so far that it is cPanel checking the up-status of Apache.
No thats not what I said.
I said if you have some sort of high load on your server the first thing everyone says is to look at the Apache status to see what the load is. How can you do this when it is filled with HTTP 1.0?
When I had Apache 2.2 my Apache status was filled with HTTP 1.0 so I couldn't see any other information. Once I down graded to Apache 2.0 I could see the traffic and the HTTP 1.0 was now at the end of the url being accessed on the server.
So it may seem that for some reason the URL half is being left out some how. Maybe not.
Oh, I see what you mean. That is a totally different possibility. I'm pretty sure that isn't what is going on here and it is more likely some "empty" request of some kind rather than a real request that is not being shown correctly.
If you have a high load caused by tons of hits to Apache, it will show up clearly in your status page. I don't think the these strange entries are going to prevent real and current requests from being shown.
I'm more concerned about what is causing them in the first place.
I agree! And Cpanel team should find a solution ASAP!
We just upgraded two of our machines to Apache 2.2 this weekend and now if I check apache status in whm I see tons of :
9-0 - 0/0/5 . 0.02 9 0 0.0 0.00 0.00 ::1 xxxxx.com OPTIONS * HTTP/1.0
10-0 - 0/0/1 . 0.00 7 0 0.0 0.00 0.00 ::1 xxxxx.com OPTIONS * HTTP/1.0
If I restart apache they start showing up in a couple minutes. Happening on both machines right after the upgrade. I'm guessing this is Cpanel itself checking status because we didnt do a whm apache status for almost a whole day and when we checked tonight there were tons of them.
It is your Apache 2.2 if you downgrade back to Apache 2.0 they will go away.
Not sure why Apache 2.2 and cpanel have this problem did anyone open a ticket about this?
Downgrading isnt an option for me, we had to go to 2.2 since we could no longer get 2.0 to compiler anymore, kept getting weird errors.
Yes I am not happy that I have to run 2.0 instead of 2.2 but I ran 2.2 2 times and both times my apache status was filled with those so until something is done I am stuck with 2.0.
I am going to do some further looking and see if Apache has any info on this if I find anything I will post it here.
Could this have anything to do with how WHM has that fairly new feature that shows the current server load? It has to make periodic accesses to get that info? Just a guess.
Possibly, but if I look at the normal apache log I see the system doing its whmstatus every 5 minutes on the dot, but the "OPTIONS *" seem to have no pattern. Sometimes I see 4 or 5 per 5 minutes other times I see more than a dozen.
It would just be nice to know what it is and if its supposed to be like that or something is going nuts and eating up the CPU.
Did you ever find luck with this? I've got a server with little traffic and a load that's higher than it should be for no apparent reason. If these checks reallly aren't hurting anything it'd at least be nice to have them removed from the apache status page.
OPTIONS HTTP/1.0 from localhost
I have changed the default cpanel message page, with a redirect to the homepage company, then i got many OPTIONS * HTTP/1.0 messages in apache status.
Reverse to the original homepage and it seems its solved.
hope it work and help anybody