The memory usage on my server is fine, I watched the CPUs for a long time and watched the server load also, the CPUs were easily riding along not often going over 20% (I have 2) yet the thing is, the server load is very strangely going mad... It's rarely under 6.0 (2 cpus) and sometimes goes into 20s, yet the server doesn't slow, nothing stops working but it simply goes red and shows it's bad news.
I've been told that the only other thing that might be causing this is the hard drives... I have 2x160GB 7200rpm hard drives - not sure if they are SATA etc..
If anyone has any ideas as to why the server load is so high - but everything else is fine -
I'm having the oddest issue. For some reason, some of the websites on my server load fine, and some take a really long time to load (2 minutes).
Now, the server load is fine, and the size of the sites aren't the issue either. I've restarted Apache and a couple more services, and still the same sites seem to load very slow.
What could be causing this since it's only effecting certain websites?
I have a problem with one of my domain. Some users around the world cannot access the website, but they can ping and traceroute the server without any problems. I am able to access the website perfectly.
Site URL is www.djluv.in
And we are using IPB 2.2
Moreover if theres another domain on the same server, they can access it.
There are about 25-30 users who are facing this problem, but from different part of the world so I cant even say ISP problem.
APF has been shutdown, no Ban Filters on IPB or cpanel.
What are the possible reasons why they couldn't access the website? how to fix the problem?
We are talking with a datacentre about colocation and im hiring a third party company to manage the servers 'on site' on our behalf.
Im looking for some feedback on the 'colocation agreement' and the 'server management agreement' to make sure were covering everything.
just to be clear, Ive got a server administrator who will access the server virtually so the only points of the agreement will be those tasks that need to be done 'on site'.
the points ive looked at so far are:
colocation agreement > 1u and 2u costs. > Server setup costs. > Power allocation and cost of additional amp's. > Bandwidth overages. > storage space for spares
onsite server management agreement > hardware replacement costs . > Hardware upgrade costs. > status monitoring. > general server management.
1) what else should be covered in the colocation details?
2) what should be included within the 'on site server management' agreement?
3) what should be included under the description 'general server management'.
for example: I think physical reboots should be part of the server management agreement, but how many on average per server per month should be included?
3) should hardware replacement cost be fixed rate per task or hourly, also do you have any suggestions regarding what to expect over a year (nothing exact, just a bit of guidance to help with the budgeting)
Combined these agreements should cover everything that we can not do remotely.
Anybody seen something like this before? We have a customer who is complaining about slow FTP speeds. At first I thought his whole connection to our server is slow (he is in the mideast, we're in the USA), but through a series of tests and screenshots I've confirmed:
1) He can download from our server via HTTP at 500KB+/sec
2) If he downloads the same file, from the same server, via FTP, it goes <1KB/sec.
I've had him try different FTP software; same result. Maybe his ISP is rate limiting FTP?
Every other customer can download fine via HTTP and FTP, and all my tests from various testing servers I have around the world show good speeds from both.
When I try to login to any accounts on my server via FTP the responce is:
Quote:
[20/03/2008 15:41:15] 331 User restore OK. Password required COMMAND:>[20/03/2008 15:41:15] PASS [20/03/2008 15:41:16] 530 Login authentication failed
I can login to their cPanel accounts fine using the exact same password so I'm not sure what the problem is. I've tried restarting the FTP server and syncing the ftp passwords using WHM -> cPanel -> Synchronize FTP Passwords but still no luck. I've tried changing the account password but I can still login to cPanel with the new password but not ftp.
I try connect to the ftp server via the IP address and domain (ftp.domain.com) but again, no luck.
i havent touched the config on my vps box for months and it has been running OpenVPN without any problems.
All of a sudden it is complaining it cant find /dev/net/tun
Nov 15 06:45:18 gizmo openvpn[23616]: Note: Cannot open TUN/TAP dev /dev/net/tun: No such device (errno=19) Nov 15 06:45:18 gizmo openvpn[23616]: Note: Attempting fallback to kernel 2.2 TUN/TAP interface Nov 15 06:45:18 gizmo openvpn[23616]: Cannot allocate TUN/TAP dev
any reason why it would stop working? i am on a VPS i have asked if the host machine parameters have been changed but in the meantime anything else i can check?
I'm currently struggling with an issue on our Apache 2.2.24 home-rolled installation on an OEL 6.2 x64 linux server that front-ends for a Glassfish cluster via proxypass.The httpd.conf manages 4 virtual hosts with each virtual host entry rewriting to HTTPS.
Anyone trying to access the HTTP address redirects to HTTPS just fine.The issue is that anyone using the HTTPS address gets redirected to the Document Root defined in Apache instead of being proxypassed to Glassfish.
I had set up a dummy DocumentRoot with a simple index.html meta redirect and what happens is that anyone directly accessing HTTPS will hit the index.html file which redirects to the the https site. At this point it simply loops to infinity.
We have installed the latest version of Plesk on our server. Our web applications work fine. But we can not send or receive emails. Neither the webmail (login failed) nor pop3 / imap and mail forms on web projects.
All the videos have been uploaded to << link removed >> and in there you will seea flash_video directory, in there should be a html file where you can double click and watch the flash video.
Now go ahead and try it, click on the HTML file and click CONTINUE and try clicking on MODULE 1.. You will see a video play to the left but on the right are bunch of POWERPOINT slides that will appear as the guy continues to speak. THAT DOESNT SHOW..
Our website is hosted by xo.com
Dont know much about the plan since its my 2nd day at the job..
But here is the weird part,
i have uploaded the samething - same exact way to another web host and it plays fine, the PPTs show up fine..
What do you guys think it is?
Before someone asks about uploading methods, i tried filezilla, coreftp and cuteFTP using both ASCII and binary methods. Samething..
I have a linux server that is only hosting one site , and although the httpd and mysql are working well , then the server load is high and when I restart the server the server load is high till the swap accelerates to 100% and then it uses the memory alot.
Anyone know of some good server load testers ( commercial )?
Im not looking for application based load testing, I need real web server load testing... need to see how much traffic this one site can take before it cries.
I have a VPS. And have had an issue both when it was 1Gig and now I recently downgraded it to 768m, because I am moving some sites to a dedicated.
However, the part I am having trouble grasping is that when I look at graphs from Munin, it will typically always show 200-400MB free memory (and free -m and top agrees with munin), but Munin shows 'committed' memory that is above the total Ram on the VPS and once the 'committed' ram exceeds the VPS limit, processes start failing.
So, why is 'committed' memory exceeding the RAM on my VPS, when Munin, free -m and top all show there is free memory available?
Code: root@server [~]# free -m total used free shared buffers cached Mem: 768 449 318 0 0 0 -/+ buffers/cache: 449 318 Swap: 0 0 0 Here's a graph that munin produces that shows the 'committed' memory exceeding the total memory. [url]
I've run into an issue lately with a few servers that I support and I'm curious if anyone has any suggestions on how to approach the issue?
Scenario:
The server will stop responding to new network connections, I can ping it, but nothing more. When the DC looks at the server, they find the error " Out of Memory " on the console.
What is a good approach to troubleshooting these?
Here is what I do, if I'm off base, I'm here to learn and will read
find / -name "core.*" -ls | egrep core.[0-9]
**try and find software failure
df -h
**validate disk usage
** check /var/log/messages /var/log/dmesg
If I can't find anything obvious in those places, I start digging through the access_logs and error_log to try and find a giant influx of requests that could have hit the server too quickly to process thus filling the memory up.
In this case, I adjust the rlimits.
*shrug* This is not fixing the issue on a few servers and I'm curious how other people approach this problem? Could this be hardware?
Since a short while back my server memory usage is 100%. The amount of visitors decreased so the memory should be lower in consumption than before.
I suspect a newly installed script could have bugs or misuse of the server by hackers. I don't know how to see on the server what script or what part makes the memory use so high. I see Apache is very high but I'm on VPS. Technical data........
Currently i have a small server with 512 Mb of ram i doing some minor hosting.. sometimes..my free memory is only left with 15 mb..is there a concern down here?
is there anyway i can find out what are using up my memory..? And should i just upgrade to 1Gb instead?
Somtimes my mail queue ..will also take up like 20 to 30 mb of memory as they keep on trying to resend themself
i'm running Suse 10 linux server in my company, and that server is used primary for graphing solutions, and data gathering.
server is fujitsu-siemens primergy rx100 s2 - suse was the only OS that detected disc controllers at that time.
It runs tons of queries to bunch of devices via SNMP and stores data in RRD files and some other data to mysql. It also runs a web server for displaying that data.
queries are short PHP script, that runs in CLI mode by cron
Few months ago my company switched to MPLS routing of our local network, and IMHO that was done very bat so sometimes i get timeouts, or networks run very slow.
other thing is that i server is behaving really weird.
Issue 1 - memory leaks or something like that After restart it works quite fine for about 10-15 days, but after that when i run top i see that apache is using more and more VIRT memory, it starts with 8 megs per process and ends up with 150-200 megs per process. things i tried to fix: recompiled from scratch (php,mysql,httpd,snmp,maybee even libc but i'm not shure about that) all latest stable version, no improvements - when it runs it runs like hell, but after few days it starts to eat memory again.
Issue 2 - strange CPU load after few days of flawless work, when it starts to eat memory it starts increase load, BUT in period of 08-20h load increases up to 100 and then suddenly drops to 1-2... and then in next morning it starts to rise again
only thing that comes to my mind is that ether:
- some computer is infected wit some virus, so in work hours it kills my machine - but ther is nothing weird in netstat
- doe tho lousy MPLS routing server is getting bunch of timeouts in work hours when we have large amounts of traffic, so the poll queue get larger and larger (cron runs every five minutes)
-everything is messed up - i need to reinstall everything (I'll rather shoot my self in the foot )
so what i'm asking here is HOW to gather data on what exactly is happening on that server, I'm quite experienced in writing shell scripts so it would not be a problem for me to write some custom scripts to analyze traffic, ports, (using iptraf) i just need someone to tell me WHAT and WHERE to look.
P.S. 1 Almost Everything on server is compiled by me, in order to get maximum from hardware. No RPM's
P.S. 2 PHP and Mysql are already well tuned and optimized - query cache works, APC is installed, for first 10 or so days it can handle A LOT of load without a problem.
We have a HP Proliant DL 360 Server, it was running with 2gb RAM, this consisted of 4 x 512mb sticks. We upgraded by adding another 4gb, this was 2 x 2gb sticks, so we removed 2 of the 512mb sticks and replaced with 2 x 2gb stickes, effectively giving us 5gb of RAM. When I rebooted the machine at the server centre it immediately displayed 5 gb of RAM as it run thru boot up, it then displayed a screen saying it had found the RAM (it looked like a BIOS type, although I'm no expert) I think if my memory serves me right it then confirmed the memory was OK and continued booting into Windows 2003. However when we viewed the maching remotely it is only showing 4gb of RAM. Does anyone have any idea why this might be? where would are missing RAM be? I guess its the 2 x 512 not displaying. Its awkward to access the server easily, it means making an appointment etc, I wondered if there is anything we can do remotely to configure the RAM.
