I have a VPS and it worked well, the problem is if I install huge database which is more than 2 gigs, the VPS will hang because it has no physical ram to handle it. My host told me the only way is to get a dedicated server for a db more than 2 gigs. Dedicated servers have physical ram to deal with such db's.
I had a 'VPSII' at 1and1 (Windows). The VPS II is advertised as having 256MB, and in Task Manager it would show physical memory as 450MB. When I first questioned this difference I was told the 450 included an 'allowance' for the OS - seemed reasonable.
I just upgraded to a VPSIII - which is advertised as having 512MB. Call me crazy, but I was expecting Task Manager to now report 706MB (450+256), but it only shows 550MB.
I called 1and1 and they said 'that's correct' - we give the smaller VPSs a 'boost' but not the VPSIII - WTF?
I live near NYC, and it would be very convenient if I could have physical access to my colo'd server. I don't suppose there is anything in the $50/month range? If not, then I guess any affordable colocation on the East Coast?
My server (cent OS4, plesk 8) was frozen for a day and the NOC had to reboot it, here is the mail I got from my host:
>Your server was frozen, with a kernel panic. Ensure that you check your logs closely to determine how this happened,
After looking at the message log here is the part of the log when the crash happened: Is this really a kernel panic, I am not sure...
Dec 8 09:05:36 server kernel: input: AT Translated Set 2 keyboard on isa0060/serio0 Dec 8 09:05:37 server hal.hotplug[2701]: DEVPATH is not set Dec 8 09:05:37 server hal.hotplug[2702]: DEVPATH is not set Dec 8 09:05:42 server login(pam_unix)[2670]: bad username [ ] Dec 8 09:05:42 server login[2670]: Authentication started for user Dec 8 09:05:44 server login[2670]: FAILED LOGIN 1 FROM (null) FOR , Authentication failure Dec 8 09:05:50 server login(pam_unix)[2670]: authentication failure; logname=LOGIN uid=0 euid=0 tty=tty1 ruser= rhost= user=root Dec 8 09:05:50 server login[2670]: Authentication started for user root Dec 8 09:05:53 server login[2670]: FAILED LOGIN 2 FROM (null) FOR root, Authentication failure Dec 8 09:05:57 server login(pam_unix)[2671]: authentication failure; logname=LOGIN uid=0 euid=0 tty=tty2 ruser= rhost= user=root Dec 8 09:05:57 server login[2671]: Authentication started for user root Dec 8 09:05:59 server login[2671]: FAILED LOGIN 1 FROM (null) FOR root, Authentication failure Dec 8 09:06:00 server shutdown: shutting down for system reboot Dec 8 09:06:00 server init: Switching to runlevel: 6 ^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@ ^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@ ^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@ Dec 9 05:52:36 server syslogd 1.4.1: restart.
It looks to me like if someone has physically connected a keyboard and logged in at the NOC.
I use Iptable to restrict ssh access to my IP each time I connect remotly, so I dont' think a remote connection has been possible.
any idea about this line: ^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@ ^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@ ^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@ I think it's just corupted data the was written when the server shutt down.
Also i didn't find any other signes of kernel panic in the logs
Looking at the httpd error log I found this lines before the crash:
[Sat Dec 08 00:44:40 2007] [error] [client 213.215.41.138] client sent HTTP/1.1 request without hostname (see RFC2616 section 14.23): /w00tw00t.at.ISC.SANS.DFind [Sat Dec 08 00:44:40 2007] [error] [client 213.215.41.138] client sent HTTP/1.1 request without hostname (see RFC2616 section 14.23): /w00tw00t.at.ISC.SANS.DFind
apparently somone doing server scan. maybe the 2 events are correlated and the server freeze could have been a result of some buffer overflow attack, but i sould be finding some evidences of this on the apache logs?
What direction should I take to investigate a bit further on this server freeze?
We have a windows 2003 machine that is currently responding properly to all services that it's running (including IIS and FTP), however RDP connections just close right away and the server isn't responding when a local keyboard & monitor is plugged in.
The last thing that I want to do is pull the power out while it's on - any ideas?
I tried a remote shutdown, but the response is: C:Users>shutdown /r /m HOSTNAME HOSTNAME: The entered computer name is not valid or remote shutdown is not supported on the target computer. Check the name and then try again or contact your system administrator.(53)
Are there any calculators, white papers, or other guidelines that provide guidance for sizing a physical server to be used for openvz or Virtuozzo for Linux?
i.e. if you want to run “x” VPS nodes totaling “y” disk space needing “z” number of processes, etc. you will need “n” CPU’s of at least ____ Ghz with ____ RAM, and ____ hard drive space of which ________ should be reserved for the operating system and openvz system software
Also for openvz or Virtuozzo for Linux, which RAID provides the fastest performance without trading off too much for hard drive reliability?
I have a Windows 2003 server with 2GB ram running IIS and SQL 2005 express. I noticed that the server only uses around 1.2GB of its physical memory (been a few days) while page file usage seems high at 1.1GB.
SQL server has been allocated a maximum 1GB of memory while IIS has no limitations on memory usage.
Server doesn't really show any slowness at this point (not a lot of traffic). But to my understanding other than the proccesses that always use some virtual memory (around 400mb usually), page file should be only used when there is insufficient physical memory? So is there something wrong with my memory usage, 1.1GB on page file while there are still almost 50% available physical ram?
I'have a problem with my aps setup on sanbox.When i create on customer ccp when i click finish i have this error. I must only test.
Error: Instance of application with id 124 and version '1-4' can not be provided: There is no resource of class 'Shared hosting Apache' with provisioning attributes 'Web Cluster' in subscription with id 1.:There is no resource of class 'Physical hosting (IIS)' with provisioning attributes 'Web Cluster' in subscription with id 1..If i add the shared hosting apache resourse i get this error : There are no "apache" services that satisfy given attributes: "Web Cluster".