Spork Boards
Hot Spork Chat : Join us in an AIM chat room!

Milkshakes for...

rino's Avatar Picture rino – December 11, 2007 04:54AM Reply Quote
Generally miffed at something? Let the world know.

Edited 1 time(s). Last edit at 12/11/2007 04:54AM by rino.

ARL (Moderator) – January 19, 2017 06:10PM Reply Quote
I whinge therefore I am!
Apple's new implementation of two-factor identification.

With some apps or devices you put the 6 digit code in a separate field but with others (such as logging into the Mac app store) you're expected to tack it onto the end of your password. This is explained far from clearly and I'm wondering if the UI bozos just thought "eh, they'll google it to figure it out".

The error message is ambiguous saying "An Apple ID verification code is required to sign in. Type your password followed by the verification code shown on your other devices" (nothing about "in the same field!!!") and then sends you on a "forgot password" wild goose chase.

To make it more alarming you get an alert to your phone saying someone is trying to login from a different address. It was only after a while that I realised this login address was the Apple data-server address, not some hacker login in at exactly the same time I kept trying to.

JW, you mentioned your Dad was getting weird error messages about a login from 100 miles away? It could be this two-factor identification "feature".

Edited 1 time(s). Last edit at 01/19/2017 06:12PM by ARL.

Cloudscout – April 03, 2017 04:10PM Reply Quote
Me, I want a hula-hoop.
I have the LG G5 with AT&T. When it's working properly, this is an awesome phone. I love the cameras. The ability to switch to the ultra-wide-angle mode is far more useful than I realized it would be.

Unfortunately, there seems to be a serious problem with the G5. In certain spots within the Twin Cities Metro Area, my phone starts freaking out. It goes from full signal to NO SERVICE every few seconds. In addition to rendering the phone unusable, it also drains the battery roughly 1 percent per minute.

It turns out, I'm not alone. Other people have reported the same symptoms. The REALLY weird thing is that ALL of the people who have reported it are in the Twin Cities as well. Nobody in other AT&T markets have reported the problem.

I've resisted calling support so far because my experience with their support staff has always been fruitless.

Tonight I went into an AT&T store with low expectations. As soon as I walked in, a guy asked what he could help me with. I said, "Well, my phone is acting up." He asked, "What's it doing?" I took the phone out of my pocket as I replied, "It keeps flipping between NO SERVICE and full signal in certain places." He looked at the phone in my hand and said, "Ahh, yes. That's a known problem with the G5. I had it with mine, too. Your only options are warranty replacement or getting a different phone. Unfortunately, a warranty replacement phone probably won't work either but you need to go through the process of getting a replacement anyway, possibly more than once before they'll do something about getting you a different device."

He gave me the phone number to call for warranty replacement. I'm going to try calling them now. I hate that I have to jump through hoops like this rather than just have AT&T take care of their customers properly up front.

YDD – August 21, 2017 04:15PM Reply Quote
Yak shaving.... and the yaks now have Rogaine.

Warning: incoming work rant, which could have gone in other places.

The past few weeks have been an interesting exercise in seeing how complicated a notionally simple task (upgrade some security monitoring software on our production systems) can be. It all starts with our nuget chain. Don't ask me why, but someone had the bright idea of having repo A push to repo B, which pushes to repo C which finally pushes to the feature repos (six or seven of them). It turns out that this change had to start in repo A. I did a quick and dirty push through, and most stuff seemed to work, but there were some edge cases to sort out.

Now repo B is a bit hairy. It's actually the original repo (prior to the imposition of largely arbitrary splitting which was introduced very rapidly and without consultation several years ago), and has been one of the last to migrate from TFS to git. In that repo, you don't ask for logic behind any choices as to layout or contents. It simply is, and you try to avoid making things worse.

When I got back to repo B last week (the TFS->git migration having occurred a couple of weeks previously), I found that the build had spontaneously broken. I'm not sure what changed, but some tests using a mocking framework had died. A little digging revealed that said mocking framework was deprecated approximately five years ago, shortly before the last of the tests were written (indeed, touched).

I then spent some time porting some tests to a supported framework, and disabling those where there were too many non-failing tests to port the whole DLL. And we eventually had a green build again. Great - push the nugets again.... and why are my roles chain rebooting on startup (before logging or anything useful came up).

One RDP later, and it turns out that the new nugets aren't getting properly signed. Just some of the DLLs in a given nuget are being missed out - others in the same DLL are fine. So I've just spent two entire days fiddling with parts of props and proj files, trying to get everything signed. This quickly degenerated into a game of Whack-a-mole with DLLs losing signing for no apparent reason - or being signed with the test key instead (equally bad). This is not aided by the debug cycle being nearly an hour long.

Part of the latter was traced to the vagaries of git and repo B itself. The full history was migrated from TFS to git, and at some point in the past, large binaries were checked in (think 20 GiB repo). And of course the VSO build doesn't do a shallow clone by default..... Even with that fixed, a test cycle is over 20 minutes thanks to the nuget chain, and that's if I can catch any issues before attempting to bring things into the feature repo.

I think that everything is straightened out now... which means that I've fixed one of the test failures in the feature repo I've been working on. Three more to go.

We now return to our regularly scheduled lurking.

Sorry, only registered users may post in this forum.

Click here to login