Quick resolution: Stop & Start the instance
For those interested in the whole story …
We had an instance go down and upon logging into the AWS EC2 Management Console, we were confronted with this scarry message
Naturally, we clicked “Troubleshoot”
The Resolution seemed simple enough – Just Stop & Start the instance; however, stopping the instance didn’t seem to work – It just hung at and instance state of “Stopping”
After waiting what seemed an eternity (~45 minute), the instance finally stopped, at which time we were able to start it back up – effectively clearing the issue of the instance being unreachable as well as its retirement.
If you’ve ever wanted to update the bandwidth limit via command-line, without needing to login to WHM, this post is for you! I was surprise with how simple this was.
- Manually edit line beginning with “BWLIMIT=” in
- Get your modification recognized by running the script below
- There is no step 3!
For the impatient, simply upgrade to at least WHMCS 5.2.2!
Per WHMCS support…
In the latest version of WHMCS we introduced a new session handling class which includes a random session name per install. We also updated the standard WHMCS verify image to use this new class and not call session start itself.
Full back-story and resolution for those not running version 5.2.2, are unable to do the upgrade, or perhaps experienced trouble with this fix.
For the impatient, simply add this line to the end of your
For those interested in the full story …
Since right around the time we installed WHMCS version 5.1.2 in July 2012, we had been receiving an “Invalid Token” error periodically, yet quite regularly, when submitting a form on WHMCS (support ticket reply, invoice, etc.).
We opened a ticket with WHMCS support in August explaining the issue, hoping for that a response that would quickly resolve the issue. Unfortunately, that didn’t happen 🙁 The support team kept saying the error was due to the PHP session expiring and that it was essentially an issue with the hosting provider. With session.gc_maxlifetime set to 24 minutes we were surprised to see the Session ID change after only a few minutes.
In March 2013, our team finally tracked down the cause of this issue …
WHMCS by default uses the PHPSESSID cookie to track its session. If there are any other PHP applications on the domain which use the same cookie, it can cause a conflict which in WHMCS results in an “Invalid Token” error. It took us a while to track this down as the other apps on the domain gracefully recovered from the shared cookie.
To resolve the issue, we added the line of code below to our
Now the session cookie is called WHMCS and does not conflict with the other PHP applications on the domain.
Did this help you? Show some love with a comment!
**UPDATED 11/13/2013 to reflect usage of Windows Azure Active Directory Module for Windows PowerShell.
- Download & install the Microsoft Online Services Sign-In Assistant for IT Professionals BETA
- Download & install the Windows Azure Active Directory Module for Windows PowerShell
(32-Bit | 64-Bit)
- Launch PowerShell and ensure we have loaded the Office365 Module
- Connect to Office365 using Administrative Credentials
$cred = Get-Credential
Connect-MsolService -cred $cred
- Set all user’s passwords to never expire
Get-MsolUser | Set-MsolUser -PasswordNeverExpires $True
To set a single user’s password to not expire
Set-MsolUser -UserPrincipalName firstname.lastname@example.org -PasswordNeverExpires $false
- Confirm users passwords are set to not expire (Optional)
Get-MsolUser | fl UserPrincipalName,PasswordNeverExpires
Thanks to Jan Egil for her great blog post which helped me figure this out and subsequently a TechNet article for the 11/13/2013 update.
I recently encountered an issue where I was unable remove a user from a Google Analytics profile. Immediate Googling was unsuccessful so I began playing around. Turns out, you cannot remove an administrator; you must first assign the “User” role to the account you wish to remove.
I was playing around with node.js on EC2 the other day and ran into some trouble installing npm. I was getting this error
$ curl http://npmjs.org/install.sh | sudo sh
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
101 7881 101 7881 0 0 94200 0 --:--:-- --:--:-- --:--:-- 148k
npm cannot be installed without nodejs.
Install node first, and then try again.
Maybe node is installed, but not in the PATH?
Note that running as sudo can change envs.
My initial attempt at finding a resolution revolved around trying to update the path env setting used by sudo. After spending some fruitless time on this I switched gears and came across Michael Dillon‘s Stack Overflow post which recommended setting up some symlinks to resolve the path issues.
sudo ln -s /usr/local/bin/node /usr/bin/node
sudo ln -s /usr/local/lib/node /usr/lib/node
sudo ln -s /usr/local/bin/npm /usr/bin/npm
sudo ln -s /usr/local/bin/node-waf /usr/bin/node-waf
Minutes later npm was up and running!
We were recently transitioned to Office365 and learned that certain settings and permissions did come across during the upgrade. This meant that some settings and permissions needed to be reconfigured via PowerShell. Back in the days of BPOS, you needed the Migration ToolKit installed in order to use any of the Microsoft Online related PowerShell Cmd-Lets. With Office365, you don’t need this! Instead, you need to open a server side connection to the cloud servers from your local machine. Below are the steps to do this.
- Load your Office365 credentials in a variable
- Create a session variable which will configure a connection to the cloud servers
$Sess = New-PSSession -ConfigurationName Microsoft.Exchange -ConnectionUri https://ps.outlook.com/powershell/ -Credential $Cred -Authentication Basic -AllowRedirection
- Initialize the session
- Run any of the Exchange Online Cmdlets found in the documentation.
- Once you’re done running commands, you can close the session
Note: The guts of this post came from a Microsoft Article which I never was able to come across in my searching, but rather was pointed to via a service request.
**NOTE** You will need to have compiler access on your shared hosting account in order for the instructions below to work. You can request your host enable this, or pass these instructions along to them and request they compile git for you.
tar zxvf git-22.214.171.124.tar.gz
echo 'PATH=$PATH:$HOME/bin' >> $HOME/.bashrc
If you get the error “bash: /usr/bin/wget: Permission denied” you can try downloading the git source with the command
curl -O http://git-core.googlecode.com/files/git-126.96.36.199.tar.gz
I recently purchased a new Lenovo T410 with an i7 processor to replace my T410 with an i5 processor. While I’m a heavy user of Dropbox for files and portable versions of many applications I use on a daily basis, I tend not to place my Virtual Machines on Dropbox … go figure :-p. This saves me a ton of time when setting up a new machine since my documents, photos, etc. are automagically copied to the machine once Dropbox has been installed. Add LAN sync to the picture and you’ve got a pretty fast way to provision a new machine
In the process of preparing my new machine, I installed VirtualBox and manually copied over the VDI and XML configuration files. Everything was looking good, until I tried starting one VM which of course had some very important software running on it.
The VM kept rebooting, again and again without stopping to show me the BSOD detailing why. Figuring it was something to do with the drivers I booted into safe mode and tried uninstalling and reinstalling VBox Addons. After rebooting, I was once again caught in the never ending reboot cycle.
I jumped back into safe mode and disabled the automatic reboot after a BSOD (See http://support.microsoft.com/kb/174630 for details on doing this) and rebooted once again, with my fingers crossed.
Success! I was finally able to see the cause of of the BSOD:
Stop: 0x000000CE (0xF7A3C8F8, 0x00000000, 0xF7A3C8F8, 0x00000000)
So now I have something Google/Bing that actually describes what’s happening. Moments later, I come across a Tickt #420 – http://www.virtualbox.org/ticket/420, which references this Stop error, but isn’t an exact match to my symptoms. At this point, I’m willing to try pretty much anything to get my VM back up and running, so I follow xok’s instructions
- Boot Into Safe Mode + Networking (F8 @ startup)
- Go to %Systemroot%\Inf\ and delete all OEM#.inf
- Search for intelppm.sys in the c:\windows\ (search for hidden directories and system files) – Delete the 3 results of intelppm.sys
- Reboot the Guest
- Reinstall VBox Additions (might need to do this a couple times to get everything functioning @ 100%)
And BOOM! I’m back alive.
Looking to resend a single SSRS subscription?
In a nutshell, subscriptions are configured via the Report Manager which then configures and are later triggered by a SQL Server Agent Job based on a Schedule ID. This schedule ID can be tied back to a report and a particular subscription owned by a user with the ReportSchedule table. From this point, you can then query the Subscriptions to see the associated User.
- Within the Report Manager, browse to edit page of the subscription you would like to manually trigger
- Locate the SubscriptionID GUID in the address bar
- Open SQL Server Management Studio (SSMS) and connect to the server with the ReportServer Database
- Run the query below replacing the SubscriptionID with what you found in step 3
WHERE (SubscriptionID = '04109b04-aaaa-bbbb-cccc-fe271bde49a4')
- Manually execute the subscription without modifying the schedule with the query below, be sure to replace the EventData parameter with the ScheduleID returned in step 4.
EXEC ReportServer.dbo.AddEvent @EventType='TimedSubscription',
Want to dig a bit deeper? Starting with these tables will give you a head start.
ReportSchedule – Relationships between Schedules, Reports and Subscriptions
Schedule – Details of all schedules
Subscriptions – Subscription details including owners, last run status, parameters, etc.
Users – Ties the Owner’s GUID to a username