Last week I found myself in a awful predicament. Slack.app for Mac would not open and would hang when trying to load a Team. I searched the Internet and then looked for the files that held the Slack Settings myself in all the usual places (~/Library/Preferences, ~/Library/Application Support, etc.). After not having any luck I even at one point tried App Zapper to see if it could find the settings folder … No Luck!
After not finding anything I had somewhat decided that the only thing to do was to contact support. Support didn’t help a lot and told me to go into the app and click Help -> Reset Local Cache. This is a great tip for most things but there was one problem. The app would not stay unfrozen long enough for me to click on Help -> Reset Local Cache.
I told the guy in support this was the case and didn’t hear back from him because of the long weekend. I finally ended up just creating another user profile and fixing the issue that way (which sucked but I was due a format / reinstall anyway so I went ahead and did that while I was at it).
Tuesday support finally got back to me and had a solution that does in fact work. That file that I couldn’t find is in fact located in ~/Library/Containers. So while this post doesn’t help me I thought maybe it might fix someone else’s problem(s) that wanted to reset their slack settings and could not. Simple go to the ~/Library/Containers folder and delete the com.tinyspeck.slackmacgap folder. This will reset your preferences and any Team Data you have in Slack. Hope this helps someone else.
So many times people ask me why something breaks on a website. They immediately want to attack the programming / programmer and sometimes forget that it takes a small village to run a website. There are also external factors such as external services, buggy client software, non-updated browsers, viruses and malware, etc. that can cause issues with websites.
The most interesting thing this chart points out is that while there is only once clear channel to move through for a success, there are several points of failure that can cause users to experience problems. One of the most common of those, which is never thought of, is that users sometimes don’t understand the scope of the task they are trying to accomplish. This causes one of two things to happen:
1) A user asks for a feature that isn’t needed.
2) A user believes a problem occurs when it doesn’t.
3) A user causes a problem because of something not being right with settings or something on the device.
With the above chart this could simplify explaining this to a customer, a boss, or anyone else that asks the questions such as “Why do things break when nothing has been changed?”.
Having a TeamViewer unattended host on a server is a million times better than remote desktop in many situations. First off screen resolution by default isn’t an issue. Secondly you can also let someone else see what is going on with the physical server. Today I came across an interesting thing. We had remote access to a server through RDP so we installed TeamViewer (Version 10). With that logged in RDP account. Everything was great and the ID for this account was quickly added into our list of servers that we manage. Then we logged off of Remote Desktop. When this happened we noticed a strange thing, the remote desktop killed the TeamViewer. After a few hours of searching the Internet to no avail and trying over a dozen things we tried locking the screen with the remote desktop. When we did we got the following error with a black screen:
“The screen cannot be captured at the moment. This is probably due to fast user switching or a disconnected/minimized remote desktop session”. That is a great bit of information but gives no solution on how to fix the problem. I thought the service must not be starting with the session so I will try a remote reboot to see if we can force the service. This is when I stumbled on the answer to the problem.
The Remote Desktop Session through RDP has a different TeamViewer ID than the standard service running. When you tell the server to remote reboot it gives you the prompt that “Multi User View Mode is enabled” and that logging on again will require you to log in with the new ID XXX XXX XXX. To connect to the correct ID note it down and then cancel the remote reboot. Once that is finished you can log out of the RDP and then log in with the new TeamViewer ID. This will be your permanent ID outside of the RDP Scope.
Lots of people looking for something like this on the Internet so I took a lot of information and compiled it together so that people can get hybrid style 2.x sites to connect to both MySQL and MSSQL without the need for a windows machine.
Not much to it folks just download the plugin and then set it up according to the git README. I’ve got a few things I have left to do on this including making some changes to the more advanced features that I haven’t fully tested but this should do for most basic things.
Interesting note after installing the patches to Debian servers for Heartbleed you need to restart any services that are dependent on SSL. Even though it says you do not, the Heartbleed bug remains until services are restarted manually.
For a while now I’ve been using Paypal Web Payments Pro. I also use the cakePHP framework almost exclusively and have written several pieces of code for it. My most recent code was a Component used to access the NVP Methods of the WebPaymentsPro API. I had released this code on BitBucket, but then decided to actually make it into a Plugin available at https://github.com/cpierce/paypalwpp-plugin-for-cakephp-2.x. I hope this helps someone out there using Paypal WPP and CakePHP 2.x. The data is sent using CURL so php5-curl support is required. It is also recommended to use an SSL Certificate when dealing with credit cards on the web.
Several clients have been getting this in the mail. The PDF attachment shows a bill owed for $65.00 to DNS Services. At first glance it looks sound enough and DNS is one of those things that seem like something you need if you have a website. The icing on the cake is that they even point out a few things that are familiar including the current DNS (ns.csdurant.com / ns2.csdurant.com in this case), Mail servers, etc. The big things to note here are the items I highlighted in green. Name Server 3 and 4 while they seem to be from DNS services show inactive. Also in the glob of fine print at the bottom you see “You are under no obligation to make any payments on account of this offer unless you accept this offer.” And finally, “DNS Services is not affiliated with your current Name Server Provider. Despite the warnings several clients have called asking if they owe this bill.
Don’t be scammed by these mailings. This information is all available to anyone publicly via a command known as whois. These people are just trying to get your money much like the fake yellow page ad’s of yesterday’s scammers. While these people aren’t really doing anything illegal they are in my opinion performing shady business practices which try to trick people into becoming their customers.
Website design has been around long enough by now that you would think there wouldn’t be so many terrible websites out there. Today I thought I’d discuss my opinions on why so many of these still exist and continue to exist. I’ll be honest with you as well, I’ve even created several of these monsters. Not necessarily willingly but sometimes what the customer wants isn’t what the customer needs and the result turns out to be catastrophic. Continue reading The reason site design turns bad.
Photos are all over the internet. Someone [or some thing] puts them there with HTML code most generally. This is usually done with a piece of code in HTML known as . The img tag as we’ll refer to it in this post is most generally used to display images on a page but often the image is left bare and raw and causes the page to look ever so BLAH. Continue reading Making images [IMG] look nice with CSS Styling