I am having a site developed that will have a "front end" - ie the actual website with a login members area etc. The "back end" will be a set of scripts that work behind the scenes, including a bot that visits other sites and passes info to those sites.
The concern I have is, if I put the whole lot on one Cpanel, and then it gets whacked by a surge of traffic etc, the "back end" will crash as well.
My question is, would it be better to host the site itself on one cpanel account, but store the Db and the back end scripts on another cpanel, both on the same server. Then even if the site cpanel maxes out or crashes, the processes will continue to run? The site itself would also need to access the Db stored on the second cpanel.
I have a couple questions maybe some of you can help me with -
I have a dedicated server and many hosted customers, I want to setup SSL for IMAP email and their cpanels.
Instead of purchasing individual SSL certificates for each domain. I was thinking to just use my hostanme (eg host.myhostname.net) and direct them to that as I have seen the option in WHM.
My question is can I use an SSL cert for email that will act as a "fleetwide" SSL cert?
For example instead of using mail.theirdomain.com could I just use mail.myhostname.com and would it give them SSL without warnings? Of course they would set their email client to mail.myhostname.com..
Is it safe or recommend to use your hostname? I read somewhere to never use your hostname as a domain.. so not sure.
I have a dedicated server, and I need to migrate a client who has with me an account with Cpanel and add on domains, and now is upgrading his account to one of my resellers accounts plans.
It has now assigned 2 dns: ns1.maindomain.com.mx IP--> xxx.xx.x.65 ns2.maindomain.com.mx IP--> xxx.xx.x.66
and he does not want to use this dns's anymore, so, he registered a new domain otherdomain.com to have new dns's I already create the dns zones for that domain and assigned 2 more IP's: ns1.otherdomain.com --> xxx.xx.xx.17 ns2.otherdomain.com --> xxx.xx.xx.18 (both different from the first ones)
The problem I have is how to separate accounts from the Cpanel add on domains and assigned them individually to a WHM, for my client to create packages and manage his own reseller accounts, without loosing any information at all.
I know that he already has a WHM with the maindomain.com.mx as his main account, but when he signs in to the actual WHM, he see just one account maindomain.com.mx
So, can you tell me how to migrate all the Cpanel add on domains to the new otherdomain.com reseller account (is already created also) with its own Cpanel each one of the 15 domains he has.
Details of my dedicated server:
WHM 11.15.0 cPanel 11.18.3-R21703 CENTOS Enterprise 5 i686 on standard - WHM X v3.1.0 Apache 2.0.63 PHP Version: 5.2.5
I have been researching the vps market for a month or so now and have started to compile a list of questions to put to vps providers who get short listed. I would love some contributions!
1) What is the cpu and how is cpu capacity distributed, by account number limits, by assigning a certain number of mhz, is the asisgned capcity burstable?
2) ram is usually clearly advertised but who scalable is it? Can you add just extra ram or do you need to upgrade to the next package. Is it burstable and with what constraints.
3) are there any limits for the number of processes (shared hosting providers may limit processes to only a few, 15 for instance before terminating them). This isnt advertsised but need to be answered for dynamic sites with high traffic.
4) Number of simaltanious connections, both from individual Ips to the sites/account or to pop3 accounts. If the pop3 account sim con is low its will be annoying when trying to donwlaod email from several of your sites at the same time....attempts after the X number will fail.
5) Will your account have assigned bandwidth or will you just be sharing whatever connection 10/100mbps with the other uses on the server. This isnt such a big deal as a lot of servers will be streched to output 100mbps of data. If the connection is a 10mbps one then its much more important.
6) if you're used to a certain type of control panel make sure they have it and at what possible extra cost.
7) Check their terms and conditions for liability regards lost data. I chose a hosting company beofre because of their superior back up system, turns out they didnt use it and I lost 5 weeks of data (about $4000 loss for me). Their t & C avioded libility for any losses inspite of the fact that they advertises the b/u facility as a special feature.
8) quiz them on "Monitoring" and "Management". Us hosting novices may see these as the same thing but hosting companies do not. Monitoring is knowing that something is wrong, management is doing something about it. Many vps providers advertise full management but wait to be asked to fix problems that could have been lossing you money for days till one of your kind users lets you know.
9) What is their infrastructure...power, location, connectivity redundency like (ie how many T1,2 or 3 do they have and is that enough).
10) Support. Is it in house or outsourced....the later is bad as they are usually given little power to do anything and you have to wait longer for an inhouse guy to get off his lazy boy.
11) Do they limit the number of emails per period (ie like 500 per hour). This wont affect some but for those of us who have large memberships to send newsletters to this is a non starter.
I think it sad on one hand..many hosts complain about users crashing/causing problems for other users on the node they are on..yet..this topic has never been covered to my knowledge.
I can fix that.
OK..the most important thing is to *NOT* try and install windows updates from Microsoft. I have heard of it crashing the entire node. Do not run Windows update either. These updates come from sw-soft and will be installed by your host. nothing for you to do here.
OK...here's the list so far of what else not to install:
I'm currently a customer of The Planet and have been now for about 3-4 years, I've been very pleased with their service and their hardware. However I'm getting to the point where I'd like to upgrade my server and their upgrade prices as like most any other datacenter are 100-200% more then retail. While I'm not against people making money, this to me seems a bit to much.
So I'm considering paying a bunch of money up front to buy a new Dell 1950 PowerEdge server, and in doing so I need to find a company that can colo the new server. However I know a little about dedicated servers but nothing about colo or where to even start. I've found a few companies here in Lexington, KY as well as Louisville, KY and a few other companies in surrounding areas but I don't know the slightest thing I should be asking about. When searching for a dedicated machine it was easy - bandwidth and system specs.
What are a good list of questions I should be asking these companies when I call them for prices and availability?
I've been hearing other admins talk about using squid to speed thins up on web servers. Yes, not as a network proxy, but as simple cache engine for dinamic sites.
My friend asked me that, "Hey! How are you doing? I am thinking about buying hosting from this new-webhosting-company-to-me.com . What do you think?"
I replied him that,
"Well, I had a server with them.. Let me tell you something about them,"
"Value for money: 4/5 points" (their hardwares are worthy) "Setup Time: 3/5 points" (they took 2 days to setup. But i read 24 hours setup time on their homepage)
"Friendly Support: 5/5" (those dudes were skill full and friendly when i asked for os reinstall and other support requests)
"Datacenter/Infrastructure: 4/5" ("good network and I got 2 hours downtime only once")
==================================
Well, I said above 4 things to consider, before buying from a host.
What would be your top 5 considerations about any host to recommend your friend?
We are too small needswise for even a half rack. Even a 1/4 rack would be overkill but nontheless, options in the area are limited to even *owned-enclosed* 1/4 racks from the colo facility itself.
The local facilities that would fit our needs spacewise are probably going to fit this in a full rack with space that we purchase against.
My concern is the security of *our* -- the customer equipment. Read alot of horror stories and would hate to end up one day finding out that whatever provider we choose was behind on bills, etc and we have X days to grab equipment from the facility, etc.
I am looking to colo SAN equipment, which is almost a triple digit box.
As per the last thread, I cleaned this up a little more, I welcome more comments as we make this the best it can be!
------------------- Top Ten Things to Consider when choosing a Datacenter
Redundant Power A minimum of N+1 power on critical systems (UPS and Generators) should be an absolute requirement for your business; however this doesn’t mean there aren’t points of failure. Not all power distribution is the same so demand a copy of your provider’s power map. 2N or greater systems is the only practical way to prevent failure. Definitions of redundant power can vary so demand to see a map that shows what it is truly redundant to. True B power should be redundant to the street.
Redundant Cooling Redundant means more than just N+1 CRAH or CRAC units. If the facility has chilled water demand either a loop feed bi-directional system or a completely redundant pipe. This allows for maintenance on the pipe without taking the system down. Other considerations include redundant chillers, pumps, valves, controls, and electrical.
Network Carriers At a minimum you should require a facility with multiple on site carriers. Competition drives pricing, therefore; by being in a carrier neutral facility with access to multiple providers, you increase your bottom line and decrease risk. Fiber should have diverse entrance paths to the building as well.
Location The risk of system outage is significantly reduced by placing your servers in a datacenter that is located in a disaster free area. The threat of natural disaster such as tornadoes, hurricanes, earthquakes, and wildfires can be easily thwarted by choosing a datacenter that does not reside in a coastal or storm centered region. Also consider the cost and availability of power when selecting your location.
Security It is important to demand accountability from your Datacenter Operator. While two-factor authentication is good, the most secure datacenters enforce three-factor authentication: something you have, something you are, and something you know. Man traps to avoid pass-back and tailgating at all points of ingress and egress should also be high on your list of requirements.
Support Do not risk your business to an unmanned facility. Require a minimum of 2 remote hands engineers and ensure the datacenter has certified professionals on site at all times. Don’t be fooled by datacenters who hire “button pushers.” Remember that your infrastructure lies in their hands during critical moments.
Flexibility to meet your business needs Don’t pay for a datacenter that is everything to everyone; in other words, avoid paying for services you don’t require. And do plan for growth, as your business grows, you want a datacenter that grows with you.
Vendors and Partners of the Datacenter Often times the datacenter operator has relationships established with vendors. Leveraging these relationships can save you time and money compared to working with solution providers.
Service Be sure to consider any other services the datacenter may offer you with regard to office space, engineering services, consulting services, customer accessibility, remote hands, etc.
Standards The datacenter you choose should be SAS 70 Type II compliant. If your business deals with online payment transactions ensure that the datacenter meets the physical and environmental controls necessary for Payment Card Industry (PCI) Data Security Standards.
I have a shared hosting account on dreamhost It has php and others installed but it allows custom installations for a number of reasons (control of the php.ini file, ...).
I have shell and ftp access to /home/username/ where my websites are (/home/username/domain.ltd)
I installed php 5.2.9 under /home/username/php5, I also installed ffmpeg but I ran into troubles installing ffmpeg-php.
Now I would like to start over, I'm not very advanced in this and I probably messed some stuff up. It seemed like a good idea to ask for some general advice before going at it again.
First, I was wondering about the filesystem I needed to set up, like the bin, tmp, lib, ... directories. Which ones do I need and what should go in them. And how exactly do I let the system know these folders are there and it should look there for some commands. When I used phpize, it used the default one instead of the one in my custom php folder. Is the export command all there is to it?
When I install PHP, where should I install it and do I need to point it to my directories (bin, lib, etc...) instead of the ones in the host's root. I noticed when installing ffmpeg-php it still looked for the default ones at some points:
HTML Code: "checking for PHP includes... -I/usr/local/include/php -I/usr/local/include/php/main -I/usr/local/include/php/TSRM -I/usr/local/include/php/Zend..." I don't know if this is normal behaviour or not, I'm just basically writing everything I am unsure about.
I got my website to use the custom PHP by editting the htaccess file
Also, what would be the best approach to remove the things I currently installed, I suppose remove the php5 directory is not enough.
I've been working in this industry for 5 years now. Over the years, I've come to realize the little things that customers do that REALLY piss tech support off. This is a guide for customers for 10 things NOT do when contacting their host's technical support team.
This is a repost of what I already posted before the big catastrophe.
Please forgive the brutal honesty. It's for your own good.
1. One ticket per issue. Emailing your issue to Support, Sales, Billing, Abuse, the owner, each individual tech, and the mayor of your town is not going to get your ticket answered any quicker. Additionally, opening 2, 3, 4, or 10 tickets isn't going to get things done any faster. Seriously - all it will do is irritate the support guy
2. Contact the proper department If your account is suspended due to non-payment, or your account hasn't yet been setup, or you want to upgrade your account - please don't bother contacting support hoping it'll get done faster. All it will do is slow down their response time to customers that have actual support issues. Billing issues goto Billing. Sales issues goto Sales. Abuse issues goto abuse. Get the picture?
3. Contact support via ONE medium If you put in a support ticket, don't get on live chat and call too. Trust me - you'll get the same answer on live chat and the phone as you will in the ticket . Same goes for requesting "updates" on your ticket - if your ticket is in queue, wait patiently for a response. If you don't get a timely response, contact the management to complain.
4. Everyone thinks their ticket is CRITICAL Tech support reps realize that you think your issue is CRITICAL and must be dealt with IMMEDIATELY. But, guess what, so does everyone else that submitted their ticket before you. Your CRITICAL ticket will be answered in the order received after everyone else's CRITICAL ticket has been answered.
5. Do not try to "bump" your ticket Making continuous replies to your ticket in an event to get a faster response won't work. In fact, in most common helpdesk applications, each reply made rotates the ticket to the bottom of the queue. So really, by bumping your ticket, you're just making yourself wait longer. Not getting service fast enough? Contact the manager of the company!
6. Include all relevant information, but only relevant information Seriously - we don't care to hear your life story. Submit your ticket with your client ID, domain name, username, password, error messages, steps to reproduce, and other information directly pertinent to your issue. If your website is inaccessible, check http://www.downforeveryoneorjustme.com/ and include your local IP address (from www.whatismyip.com) and a traceroute. That will save you a reply.
7. Just because YOU can't see the website does NOT mean the server is down So please - don't come shouting at us claiming we're fraudsters and have horrible uptime and demand a credit. Most of the time you will find there is either a firewall issue or a routing issue - or scheduled maintenance. Check http://www.downforeveryoneorjustme.com/ and your host's forums before screaming at them.
8. Avoid live chat & phone support Unless you have a quick question, live chat and phone support are probably not going to be good avenues. Chances are, if your issue requires someone to login to the server to investigate, you're just going to be escalated to a support ticket. Instead of whining about how long the support ticket will take to get answered - just get it in queue. Figure if you spend 5-10 minutes on the phone only for them to tell you that you need to submit a ticket - that's 5-10 minutes that your ticket could have been looked into. Think about it. If you do call or chat - be brief - and keep in mind we have other customers to help.
9. We don't make the rules If you don't like a company's policies or procedures, don't complain to your support tech about it. They don't make the rules, they just follow them. If you want a change, contact the management of the company.
10. Do NOT disrespect or mistreat support people If you curse at us, disrespect us, or mistreat us in any way - you can almost be guaranteed that we won't be going out of our way to help you beyond the minimum. By polite, cordial, and courteous to your support tech and it will get you a LOT farther. We don't get paid enough to deal with people's abuse.
11 (Free bonus ). The amount of money you pay does not matter to us Seriously - the fact that you pay us $9.95/month does not matter to us. We're going to provide you with the same support that we provide somebody that's paying $3.95/month or $99.95/month. Don't expect better treatment based on the amount of money you pay.
I've been working in this industry for 5 years now. Over the years, I've come to realize the little things that customers do that REALLY piss tech support off. This is a guide for customers for 10 things NOT do when contacting their host's technical support team.
Please forgive the brutal honesty. It's for your own good.1. One ticket per issue.
Emailing your issue to Support, Sales, Billing, Abuse, the owner, each individual tech, and the mayor of your town is not going to get your ticket answered any quicker.
Additionally, opening 2, 3, 4, or 10 tickets isn't going to get things done any faster.
Seriously - all it will do is irritate the support guy 2. Contact the proper department If your account is suspended due to non-payment, or your account hasn't yet been setup, or you want to upgrade your account - please don't bother contacting support hoping it'll get done faster. All it will do is slow down their response time to customers that have actual support issues. Billing issues goto Billing. Sales issues goto Sales. Abuse issues goto abuse. Get the picture?3. Contact support via ONE medium
If you put in a support ticket, don't get on live chat and call too. Trust me - you'll get the same answer on live chat and the phone as you will in the ticket . Same goes for requesting "updates" on your ticket - if your ticket is in queue, wait patiently for a response. If you don't get a timely response, contact the management to complain.4. Everyone thinks their ticket is CRITICAL
Tech support reps realize that you think your issue is CRITICAL and must be dealt with IMMEDIATELY. But, guess what, so does everyone else that submitted their ticket before you. Your CRITICAL ticket will be answered in the order received after everyone else's CRITICAL ticket has been answered.5. Do not try to "bump" your ticket
Making continuous replies to your ticket in an event to get a faster response won't work. In fact, in most common helpdesk applications, each reply made rotates the ticket to the bottom of the queue. So really, by bumping your ticket, you're just making yourself wait longer. Not getting service fast enough? Contact the manager of the company!6. Include all relevant information, but only relevant information
Seriously - we don't care to hear your life story. Submit your ticket with your client ID, domain name, username, password, error messages, steps to reproduce, and other information directly pertinent to your issue. If your website is inaccessible, check [url] and include your local IP address (from www.whatismyip.com) and a traceroute. That will save you a reply.7. Just because YOU can't see the website does NOT mean the server is down
So please - don't come shouting at us claiming we're fraudsters and have horrible uptime and demand a credit. Most of the time you will find there is either a firewall issue or a routing issue - or scheduled maintenance. Check [url]and your host's forums before screaming at them.8. Avoid live chat
I've got a problem with an SQL query that is behaving badly. Its probably really obvious, but I can't see the problem!
I have a table with 3 fields:
Code:
money bigint(20) income int(12) userID mediumint(7)
I have a row that has a negative money and a value of 0 for income. When I update the money value to add "0" to it, nothing happens (as expected). When I update the money field to add the income field (which is equal to 0) to it, MySQL flips the sign to make the money field positive.
The following queries show the problem:
Code:
mysql> SELECT money, income from users where userID=327961; +----------------------+--------+ | money | income | +----------------------+--------+ | -9223372036854775807 | 0 | +----------------------+--------+ 1 row in set (0.00 sec)
mysql> SELECT money, income from users where userID=327961; +----------------------+--------+ | money | income | +----------------------+--------+ | -9223372036854775807 | 0 | +----------------------+--------+ 1 row in set (0.00 sec)
mysql> UPDATE `users` SET money = money + income WHERE userID =327961; Query OK, 1 row affected, 1 warning (0.00 sec) Rows matched: 1 Changed: 1 Warnings: 1
mysql> SELECT money, income from users where userID=327961; +---------------------+--------+ | money | income | +---------------------+--------+ | 9223372036854775807 | 0 | +---------------------+--------+ 1 row in set (0.00 sec)
mysql> Notice that there is no negative sign after the third select!!
So, we are saying that executing an update where 0 is specified in the query results in the correct action but if you take the 0 from another field it takes the negative sign away...
This for various reasons is a rather urgent problem. This problem has only appeared after upgrading from MySQL 4.0 to MySQl 5.1. This problem does not occur in MySQL 4.1.
I have a dedicated server. It is a Fedora Core 6. It came with "Plesk 30 domain license". However, this license of Plesk doesn't have SpamAssassin enabled. In order to get the license with SpamAssassin, it costs an extra $30 a month.
I believe SpamAssassin by default is in fact installed on the sever, but some of my users are reporting that they are getting Spam. How can I check is SpamAssassin is running? How do I configure SpamAssassin to filter spam on all the e-mail addresses that are created in Plesk?
how to install CentOS by not using the installer. This guide should be great when installing over networks, don't have a graphical console available (for installing over serial), when you're not content with the installer's job, installing CentOS from another distro, or plainly want to learn more about how CentOS works.
Requirements: * Have a host OS that has the "rpm" package manager available. Some distributions have it in their repositories (even if the package manager for the distro itself is not rpm), and knoppix (a linux live/rescue CD) has it aboard too. You can use the first CentOS ISO CD too (use linux rescue at boot), and it has all the necessairy packages aboard * Access to the CentOS base repository. It's on the first CentOS ISO CD * Use your BRAIN. This guide is ment to be interpreted, not copy/pasted
Code: # First, setup your disks to your liking. You can use whatever you want here, # RAID, LVM, etc... Remember your disk configuration because you'll need it # to configure grub, menu.lst and fstab. Using RAID, LVM, or others will require # more configuration than this guide covers. To keep it simple I'm using a # single disk. An example:
$ fdisk /dev/sda $ mount /dev/sda3 /target $ mkdir /target/boot $ mount /dev/sda1 /target/boot
# Depending on the host OS you're using, you may need to initialize the rpm db # on the host OS $ rpm --initdb
# Use the following command to install the packages. I'll be addressing this # command as $rpm.
$ rpm --root /target -i
# Use your shell's tab completion to complete the package filenames. I # deliberatly left out the versions so these instructions apply to a wide range # of versions
# Let's install some basics $rpm setup basesystem filesystem
# Install bash first, this is needed for post-install scripts $rpm bash glibc glibc-common termcap libgcc tzdata mktemp libtermcap
# Install some dependencies (this is mainly to keep the next command smaller) $rpm grep pcre libstdc++ info ncurses zlib gawk sed ethtool
# You may also want to install your favorite editor $rpm nano
# This provides /root with some defaults, like color highlighting on `ls` $rpm rootfiles
# Right now you have system which you can chroot to, so we can start setting up # the basics
# Mount directories for chroot operation $ mount --bind /dev /target/dev $ mount -t proc none /target/proc $ mount -t sysfs none /target/sysfs $ chroot /target
# This constructs /etc/shadow $ pwconv
# Configure fstab $ nano -w /etc/fstab
# Installing the kernel. Do this back outside the chroot in the host OS system $ exit $rpm kernel mkinitrd cpio device-mapper-multipath dmraid gzip kpartx lvm2 nash tar less device-mapper-event
# Install the bootloader, grub. $rpm grub diffutils redhat-logos
# Let's chroot again to configure our bootloader $ chroot /target
# We start by configuring the bootloader. Open /boot/grub/menu.lst, and put the # following there
# If this command gives an error, you can safely ignore this because it's not # of importance. What is important is that grub-install copied the right files # to /boot/grub that we need for booting. $ /sbin/grub-install /dev/sda
# Manually install grub if the previous step failed. - means type it in the grub # shell $ grub $- root (hd0,0) $- setup (hd0)
# Optional packages # You may want to install passwd so you can set passwords ;-) $rpm passwd libuser openldap cyrus-sasl-lib
# These are used to set the keyboard language (loadkeys) $rpm kbd usermode
# ** Right now you should have a bootable system! Here are some tips to help you # through your 1st boot ***
# Most of the system configuration happens in /etc/sysconfig. See
/usr/share/doc/initscripts
for full documentation.
Some quick post-install tips: * Configure your keyboard in /etc/sysconfig/keyboard using the KEYTABLE variable
* Configure networking Take a look at /etc/sysconfig/network-scripts. See ifcfg-lo for an example.
# This recreates the RPM database. If the host OS you used has a different # version of db, rpm will complain with # rpmdb: unable to lock mutex: Invalid argument $ rpmdb --rebuilddb
we're using a subdomain to point to one of our server's IP. (gaming purposes).
And people use this subdomain to connect to the game server. However, We are going to move to a new server soon with a different IP.
I know, only thing I have to do is change the IP of the subdomain to point to the new server, however I know this will take like 1 to 48 hours to fully work.
Is there a way to force people who's still connecting to the OLD Ip to go to the new IP?
I have put an Access database inside an access_db folder on Godaddy and written some .asp pages that query it. I am trying to make sure that I take necessary precautions against hackers reading or even writing to the database. Maybe someone can give some remarks about whether any of these concerns are realistic, and if so, why and what I could do about it?
1) Could someone somehow navigate directly to the database and read or write to it (the access_db folder seems to have no read/write permissions as set by default by Godaddy, but how secure is that?)
2) I permit entry through use of a a userid and password that are looked up in an mdb in the same folder (not listed in the html itself). If there’s a match, I store the userid as a session cookie. Then, to visit any other pages, each page first checks to see if the cookie is empty before proceeding. Is it possible for someone to set the cookie themselves and thus break through (can a cookie be set manually?) If so, would it help if I mandated that the cookie be set to something specific (right now it just has to be non-blank) or can they find out what the cookie should be set to as well?
I noticed, that my Plesk install still runs an older Roundcube Installation (somewhere in the 0.9 range) and RoundCube 1.0.3 is currently the latest version available at [URL]
Is it safe to upgrade the Roundcube Installation in psa-roundcube manually from 0.9 to 1.0.x?
I'm trying to create a script to archive logs for 7 days but still delete them from the domlogs daily. Has someone already done this? The ideal solution would be to modify the CPanel script that deletes them after stats run but I'm not sure if thats protected code by CPanel. The other option would be to disable the delete logs after stats run option and to create a script to copy the logs somewhere else via cron and call the script that CPanel uses to delete the logs without restarting apache. This script would also move the files through directories and eventually delete them, therfore preserving the logs for 7 days. Any ideas?
I have created an untested script which should do this but I just need to know how CPanel deletes the logs without restarting apache.
