![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
I got woken up by another after-hours call. It wasn't so bad - came in a bit before 8, and I'd already stirred once and just decided I didn't want to get up yet. The client deliberately waited to call until later on, because while it was important, it wasn't super-critical.
The issue with no evidence yesterday, except he got my message after removing the evidence from it happening again. Which I'd half-expected. That's okay, I'll get the next one. But in addition, he had two people who couldn't log on, and thought there was some sort of data corruption going on.
I tried to log on as them, read the error messages, and realized the poor guy had never been trained about operator lockouts (you know, try your password too many times, it locks you out) because he was trained prior to that being added to the system. (The site had hardware problems which resulted in a longer delay from training to live.) Oops.
So I showed him the commands (which are really simple) for checking and then clearing lockouts, and all was good. Also mentioned the value of the lockout metric (number of failed logins before we lock it); he asked me to change it, so that will be in their next build.
As after-hours calls go, that one was nice and easy to resolve, at least.
The issue with no evidence yesterday, except he got my message after removing the evidence from it happening again. Which I'd half-expected. That's okay, I'll get the next one. But in addition, he had two people who couldn't log on, and thought there was some sort of data corruption going on.
I tried to log on as them, read the error messages, and realized the poor guy had never been trained about operator lockouts (you know, try your password too many times, it locks you out) because he was trained prior to that being added to the system. (The site had hardware problems which resulted in a longer delay from training to live.) Oops.
So I showed him the commands (which are really simple) for checking and then clearing lockouts, and all was good. Also mentioned the value of the lockout metric (number of failed logins before we lock it); he asked me to change it, so that will be in their next build.
As after-hours calls go, that one was nice and easy to resolve, at least.
