My friend is setting up a media sharing site which will use a lot of bandwidth. He would like to colo two 2U servers with 10TB or so of bandwidth each. I have advised him that FDCservers unmetered is about the best price he can have. Can anyone else suggest a place offering 100mbit or 10TB of transfer for less than $1000/month per server.
View 14 RepliesBest management with Lots of bandwidth...
I am going to be starting a filehosting site soon and i need a dedicated server fully managed one. With that i also need alot of bandwidth.
Budget: $150-200
I was looking at [url]but didnt like it much.
I've a few arcades sites and they're using more bandwidth than my servers can handle (> 2TB month). I currently have two dedicated boxes.
The sites are static, so they don't require much processing power, just bandwidth (lots of .swf files).
My current host(liquidweb) offered me a 10mb unmetered for 100usd/month, which sounded too much, as I think that I'd be better renting more servers(Or VPSs) and spreading the sites (avoiding single point of failure).
I searched and thought about trying mediatemple grid service, which offers 1TB bw/month.
What do you guys think? Will I be able to use all the BW they promise, ow they will try to kick me off saying that I'm using too much resources(even I hosting only static files)?
I usually move 900MB to 1.1GB or so of Data per month.
I noticed today that I've moved 3.5GB this month. How can I tell what file(s) are being moved?
I'm uncertain how I have tripled how much data I usually move.
I use a host that gives me way more bandwidth and space than I need, but it is still super slow. It's Dreamhost who is known for charging very low fees and promising the world. Although my sites do load and I get unlimited everything pretty much, I can't stand the slow loads.
They have offered me virtualization of my server so I can create my own virtual server with 150Mhz and 150MB of dedicated CPU/memory for $15 more a month... do you think that will speed me up a lot or should I just suck it up and pay for a better host that doesn't pimp themselves out so badly?
I run a large service on 2 machines (we're talking 8x2.66Ghz processing power, 4-6GB ram, etc), however we are thinking about creating a node/clustered setup to help spread the load, bandwidth, mysql (having one server for hundreds of thousands of databases is not wise it seems :p), diskspace etc. Lots of machines just seems to be a cheaper and more powerful option.
Would you advise going with lots of lower end machines, or just a few mid-range machines.
Here is what I propose:
3 of these
Dual Xeon 3GHZ Dual-Core (4x3GHz)
3GB RAM
500GB Sata Drive
4000 GB Bandwidth
100mbit
or
4 of these
XEON X3220 QUAD CORE 2.4GHZ 1066FSB
2GB DDR2 RAM (Upgradeable)
250GB HDD
2000GB Bandwidth
100mbit
or
7 of these
Dual Xeon 3.06 Ghz
2GB DDR2 RAM
250GB SATA
2000GB BW/MONTH
100mbit
All of the above are roughly similar prices in total.
We are running an application on IIS that sends out 4-500 emails per day via a smarthost, but nearly 11,000 every weekend. None of this is bulk mail...
We need a new smarthost provider ASAP, as our current host is discontinuing this service this week, and we were just notified.
Our current problem is that when we try to start sshd daemon we get the error "/lib/libc.so.6: version GLIBC-2.4 Not found".
The next problem is that when running things such as "hostname" we get the error "Floating Point Exception (core dump)."
OS: Fedora Core 4
Kernal: 2.6.12-1.1398_FC4smp
Control Panel: cPanel 11.3
Think it could be cPanel? If so, how to downgrade?
Also, when I do ls for the / directory, I see a whole bunch of core.***** files.
know a good VPS hosting service which offers a lot of RAM & Port speed for a cheap price?
Best one I found so far at the "VPS Hosting Offers" forum:
Code:
Windows 2003 Enterprise
2048 MB RAM
20GB Space
1000 GB Bandwidth
$70 dollars USD
This would be used for hosting a gameserver~
I have been experiencing a lot of Keepalive requests for a particular image on a particular domain. please see the lines below.
0-11233931/63/63K 0.15100.40.030.03 195.68.185.13mydomain.comGET http://mydomain.com//images/logo.jpg HTTP/1.1
1-11233941/77/77K 0.18100.40.050.05 122.164.58.63mydomain.comGET http://mydomain.com//images/logo.jpg HTTP/1.1
2-11233951/42/42K 0.76000.40.170.17 89.139.214.74mydomain.comGET http://mydomain.com//images/logo.jpg HTTP/1.1
3-11233971/57/57K 0.04000.40.020.02 82.199.98.229mydomain.comGET http://mydomain.com//images/logo.jpg HTTP/1.1
4-11233981/46/46K 0.27000.40.040.04 217.150.55.41mydomain.comGET http://mydomain.com//images/logo.jpg HTTP/1.1
These are just a few lines from the top.
How can i prevent this from happening.. it seems as a SYN Flood, or maybe a DDoS.
I have a project (using multiple servers) that has generated 3TB of data (compressed), with another .5TB or so each month, and the hard drives on my servers are very close to capacity.
The data doesn't need to be accessible online, so I'm trying to find a low cost way to deal with this data. Keeping the data on dedicated server(s) raises the monthly cost every few months, which isn't good. Transferring that much data to my location via the Internet is costly and inconvenient.
I'm thinking the best option would be to find a host that can ship hard drives to me, after I copy data over. Does anyone know if that is something that hosting companies are willing to do?
A weird thing happents once on few days on my server, and it's not a regular thing or on exact time.
When I have this problem no page can be loaded in the browser, but WHM is working, i think because it's accessed by IP .
In apache status page I can see lots of ..reading.. requests which are there even for 10 - 20 seconds sometimes.
Usualy the server has 10 - 20 requests/s :
CPU Usage: u35146 s2297.05 cu2.74 cs4.5 - 6.97% CPU load
11.2 requests/sec - 83.0 kB/second - 7.4 kB/request
10 requests currently being processed, 8 idle servers
But when the ..reading.. requests appear it goes much higher like 100 to 200
11.5 requests/sec - 85.3 kB/second - 7.5 kB/request
200 requests currently being processed, 0 idle servers
after 3 minutes :
160 requests currently being processed, 30 idle servers
When I logged on to SSH I saw that there are ~150 conection from a single IP .
1) What would you guys say is average in terms of the # of emails in the Mail Queue?
2) What steps can be taken to tighten it up? If I start with a clean slate, it only takes about a week or less for my mail queue to reach 1000 or more. Most of it seems like junk mail.
We are going to offer web space to our students and we want to build a new server just for this purpose. We were looking at buying a storage array, a RAID card, and a server. I have no problems building a server; however, I have never built anything with an external enclosure. I am thinking about buying this storage array:
www newegg com/Product/Product.aspx?Item=N82E16816702010
I plan on installing hard drives in every drive bay for a total of 12 drives.
The problem I'm having is I don't know what type of RAID card to buy. I saw this one; but, I don't know if it will work with this storage array.
www newegg com/Product/Product.aspx?Item=N82E16816116062
Thanks so much for any light you can shed on this! I apologize for the link but I do not have enough posts insert URLs.
I have anywhere between 80,000 - 90,000 webpages that have a single code into this. Unfortunately at the time the web developer I used didn't use PHP includes. So each .html file has the code in it.
I want a way so I can do a single command either a program or an SSH command can find the syntax in the files and replace it will code I have. Its just a single line of code that is basically for an adsense code, so the pub-blahblahblah etc..
Does anyone know of a SSH command I can use, or a program that will find and replace without manually opening up each file? 80k - 90k of opening files then find/replace will take forever!
We used Red Hat with ext2 as our file system on an old server with 100k+ of image files in a single directory. This seemed to preform ok until we switched to FreeBSD using UFS. Now images load very slow. I have read that Ext2 uses and internal hash to speed lookups, while UFS does linear searches for lookups.
View 2 Replies View RelatedI have installed APC on our Windows server. After I enable APC and restart apache our pages throwing Lots of PHP Notices.
Code :
System Windows NT 6.1 build 7601 (Windows Server 2008 R2 Web Server Edition Service Pack 1) i586Architecture x86Server API Apache 2.4 Handler Apache LoungeThread Safety enabledPHP Extension Build API20100525,TS,VC9Apache Version Apache/2.4.2 (Win32) OpenSSL/1.0.1c PHP/5.4.4Virtual Server Yes
I have a Plesk v12.0.18_build1200140606.15 os_CentOS 6 server. using postfix
And Lots of spam is send from my server.
I tried: [URL] .... but with no results.
on the moment the /var/log/maillog file is over 5,5GB
and the /var/log/maillog.processed is over 7,2GB and splitted into multiple .gz files
What can I do to find the source of the problem and stop it ?
Does anyone know of a hosting Linux package (must be UK based) that has the usuals (PHP, MySQL, subdomains, email, Apache ModRewrite) that is geared towards hosting lots of low bandwidth sites?
I use several great hosters but they limit the amount of addon domains or charge you through the roof for extra ones. I'm thinking a package that will let me do 15 - 25 domains. More would be a bonus. The bandwidth allowance is not a problem. A lot of my customers' sites use less than 100 MB a month.
I have a Sonicwall NSA 3500. Does anyone know if you can use it to create bandwidth charts of Network Objects (e.g. IPs or Groups of IPs)?
I am thinking something like Cacti, but I don't know the code to pull that info or if it's even technically possible.
I have a list of a few thousands subscribers and want to use a joomla component to handle my sign ups and announcements. Problem is my web host only allows a certain amount of emails per hour, and a cap on emails per day. Are there any email hosts, meaning, a service I can sign up with that will allow me to send these emails? I don't want anything complex like aweber and constant contact. I just want the articles on my site's news section to become the actual announcement/newsletter. I thought about Feedburner, but they don't let you import your subscribers.. I wouldn't be sending announcements every day, but perhaps a couple of times per month..
View 24 Replies View RelatedI have a cPanel server with 4GB RAM
My server hangs time to time, once or twice a day. This is the last status when it happens.
Server Load 75.90 (2 cpus)
Memory Used 92 %
Swap Used 65.78 %
When I was still inside SSH when it happens, the processes I can see inside "top" are a lot of httpd processes.
So I "killall -9 httpd", I had to do it 30~40 times
#killall -9 httpd
#killall -9 httpd
#killall -9 httpd
x 30~40 times until no pid process found & the server load is back to normal.
Before that, I check httpd fullstatus, they look very normal, same goes to what I see inside WHM Apache status & cpu usage status.
At first I thought it was a memory problem after consulting with a server admin, so I replaced all 4GB new ram stick (such a waste)
which case is more preferred: a shared web hosting service with unlimited space/bandwidth, or a dedicated one with limited space/bandwidth?
View 4 Replies View RelatedHey can anyone tell me How the whole bandwidth thing works? like lets say..if i get 100k views a month, how much bandwidth would that take?
View 7 Replies View RelatedHow will I assign the bandwidth for a VE in virtuozzo power panel. I could not find any fileds that corresponds to bandwidth in the steps during creation. I could find how to restrict memory(vmguarpages) and disk space. But where will I assign the bandwidth that a VE can use.
Also what is this bandwidth rate 8kb/s or so?
Now I keep very close tabs on my site stats. Just tonight one of my sites suddenly showed 167gb for "traffic not viewed".
That was a jump from 5gb over the course of a day.
That was done in 604877 pages and 607138 hits.
Can someone explain to me what "traffic not viewed" actually is within AWSTATS.
Next I located where most of the bw went and it appears to be http code 206 showing 159gb. My latest visitor report for the addon domain only showed one ip with an odd couple direct requests. The referring site was my own cpanel but registered to a ip in India.
How can i locate where the leak is, latest visitors isn't being much help, I have since denied the india ip which i suspect was the problem maker. How can I prevent instances such as this in the future? Being a streaming video site I have the bw to spare but it is still a bit un-nerving.
Now on top of everything Virtuozzo isn't showing anything of the sort, what do i believe?
I like the LSN server company and have a server there that I plan to keep forever, they do a great job just that I am confused about their bandwidth. Also the prices aren’t exact!
Just wondering what everyone else here thinks about this:
So... LSN are doing a Q9300 server at the moment for $130 “ish” with 2TB of bandwidth. Now let’s say someone wants an extra 0.5TB of bandwidth they pay around $50 for it and if they want 1TB they pay about $95.
So you buy a server for $130 and get 2TB of bandwidth and also the costs of the hardware, space, power is covered for the hardware. Yet they charge nearly the same price for 1TB extra to a single server, which uses no extra space, hardware or power. Another thing is they charge $25 server to pool the bandwidth between your servers.
So what I see:
2 Servers 4TB B/W @ $260 + $50 (pooling charge) = $310
1 Server 3TB B/W @ $230
1 Server 4TB B/W @ $300
So as you can see from the above you can get twice the space, RAM, CPU and Power for only $10 more? Surely it costs them more than the $10?
Also my final point is LSN have a private network, that is not bandwidth monitored and your servers can use, I asked LSN if you could tunnel the traffic from one server to another, their answer was “yes if you have the technical knowledge to do so”... Well that would get rid of the $50 pooling costs... making it actually alot cheaper to get two servers.