Dear Users... (A thread for Sysadmin, Technical Support, and Help Desk people)

Status
Not open for further replies.
It's okay. We're all beat by Robert Barker and Martin Lucas, the Royal Printers in London who in 1631 omitted the word "not" from "thou shalt not commit adultery" in the Book of Exodus in their printing of the King James Bible.

They got called before the Crown and the Archbishop of Canterbury to explain themselves, fined about 50,000 dollars in modern money, and had their printer licenses revoked.
 
Last edited:
JoeMorgue said:
It's okay. We're all beat by Robert Barker and Martin Lucas, the Royal Printers in London who in 1631 omitted the word "not" from "thou shalt not commit adultery" in the Book of Exodus in their printing of the King James Bible.

They got called before the Crown and the Archbishop of Canterbury to explain themselves, fined about 50,000 dollars in modern money, and had their printer licenses revoked.
Click to expand...

Ah, yes, "the Wicked Bible". Caused quite a stir over such a simple mistake.

While the previous year there had been a similar error with "the Bizarre Bible" which accidentally replaced the entirety of the Book of Daniel with the words "lettuce hippopotamus all that she wants is another baby she's gone tomorrow ninja crocodile" repeated over and over. It took fifty years before anyone noticed.
 
TragicMonkey said:
While the previous year there had been a similar error with "the Bizarre Bible" which accidentally replaced the entirety of the Book of Daniel with the words "lettuce hippopotamus all that she wants is another baby she's gone tomorrow ninja crocodile" repeated over and over. It took fifty years before anyone noticed.
Click to expand...


They still haven't noticed a similar mistake for Revelations.
 
Wudang said:
They still haven't noticed a similar mistake for Revelations.
Click to expand...

Sigh. [pedant]That's not the name of that. It's not "The Book of Revelations". It's "The Book of Mega Star Crystal Princess Megumi-chan and the Magical Robo Knights".[/pedant]


(It's just "Revelation", not plural.)



In the OVA, Megumi-chan becomes a vampire because she unified her Star Crystal with the Galaxy Fountain to defeat the Wizards of the Dead. It's worth watching because they switch up the relationships so Akiko is dating Hideo and you know that means wacky antics!! *catface*
 
arthwollipot said:
arthwollipot said:
Yeah, we're going through the process of migrating to Office 365 right now. It's a pain, but I do believe that it will be better when the process is complete.
Click to expand...

Now that the process is more or less complete, it definitely is better in a number of ways. It's faster and less prone to random errors, and Outlook doesn't seem to drop into Working Offline for no apparent reason any more.

The transition has been pretty tricky, but I think we're better off for it.
Click to expand...




We were using G-mail but moved to Office 360 some years ago. Unfortunately our G-mail documentation wasn't transferring very well. Fortunately, our customer, at this site, was using G-mail and since that documentation was mostly reporting for them and we all had been given customer G-mail accounts. We migrated that documentation to the customers G-mail site. Now I have two work E-mail accounts and even two work computers. As part of the customers system security is that their accounts and sites should only be accessed on their equipment. So work from my company, that i create on our computer, I have to E-mail to myself on their E-mail site to print out on their printers from their, provided to me, laptop.
 
I just spent about nine hours (on a Saturday!) to figure out a weird variance between two SQL queries that should have been returning identical results. Turns out that, for whatever arcane processing reasons, if you happen to be doing a partition by window function to assign row numbers and order it by multiple columns and specify descending for one of them, you then have to specify ascending for the others instead of just leaving it assumed because it may decide to just do that second column descending, based on whatever it feels about the rest of the query.

It only affected 16 results out of 340,000 and nobody but me would have even noticed, but god damn I could not relax until I figured out why it was happening! Or rather, how to fix it. The why of it has something horrible to do with "parallelism" and the workings behind SQL processing. I don't go too deeply into that stuff because it hurts my brain and drives me mad. MAD, I tell you!

Thankfully it's such a blastedly niche situation it will be unlikely to ever occur again in my work. Still, if it does, now I know! Hooray!
 
TragicMonkey said:
... if you happen to be doing a partition by window function to assign row numbers and order it by multiple columns and specify descending for one of them, you then have to specify ascending for the others instead of just leaving it assumed because it may decide to just do that second column descending, based on whatever it feels about the rest of the query...
Click to expand...
There is no guarantee of implicit ordering in SQL. If you care about or depend on order you must specify it. It's also generally bad practice to depend on sorting by clustered keys, your DB admin should be able to change that in any way they think will benefit performance.
 
RecoveringYuppy said:
There is no guarantee of implicit ordering in SQL. If you care about or depend on order you must specify it.
Click to expand...

I know that, for choosing which columns to order by, but I've never before had an issue where I wrote "order by field x" and it decided sometimes to order it descending; I thought the default behavior was to order fields ascending unless told specifically DESC.
 
Friday night something changed in our UAT database causing the same table load job to fail in two of our overnight batches (it probably failed in the third one, but no one cares about that one at the moment). The DBA's (contractors working at our mainframe service provider) tried something, didn't test it, instead they hoped that Saturday night's batches would run OK. They didn't.

Now, they've said that they thought the changes they made were dynamic, but maybe weren't, and have their fingers crossed that tonight's scheduled recycling of the databases will fix the problem.

I am annoyed
 
TragicMonkey said:
I know that, for choosing which columns to order by, but I've never before had an issue where I wrote "order by field x" and it decided sometimes to order it descending; I thought the default behavior was to order fields ascending unless told specifically DESC.
Click to expand...


Oh. I had a read of the SQL 92 standard (as that's what I continue to work with) and it says
3) If an <order by clause> is specified, then the ordering of rows of the result is effectively determined by the <order by clause> as follows: a) Each <sort specification> specifies the sort direction for the corresponding sort key Ki. If DESC is not specified in the i-th <sort specification>, then the sort direction for Ki is ascending and the applicable <comp op> is the <less than operator>. Otherwise, the sort direction for Ki is descending and the applicable <comp op> is the <greater than operator>.
Click to expand...

http://www.contrib.andrew.cmu.edu/~shadow/sql/sql1992.txt

I may need to reread that a couple of times to make sure it says what I think it say.
 
Wudang said:
I may need to reread that a couple of times to make sure it says what I think it say.
Click to expand...
I think it says what TM expected. And what TM expected is what I would expect now that I understood he only assumed the default order and not implicit sorting.
 
I'm more of a practical worker than an expert in theory-- I tend to just try things until they work, and verify it from looking at the base data. In this case there was a lot going on, way too much grouping by one field while sorting by another to get the combination of a third field's data in light of what a fourth field contains... I regard overly-complex queries like prescription medication: the more you throw in there, the more likely there's going to be some interaction you don't want. I prefer to keep it simple, even if that's less elegant and requires grunt work in Excel afterwards. (I like to dump the raw data into Excel and do pivot tables, so if anyone questions my numbers I can just point to the actual detail underlying everything. Couldn't do that here because there was just too much data.)

I'm not actually expected to be a SQL guru in this job -- we have a much more technical team for the real stuff, I'm one step away from them closer to the business side of things. They know more about how the databases work, but I know more about what the data in it actually means and can be used for. So I'm not a complete failure. But I'm definitely not an expert in this crap, even after twenty years of playing with it.
 
Today is going to be the last time I'll be working from home. During COVID our policy has been that if your shift ends earlier than 4:30 it's done in the office, and if after then, from home. So I've been doing a lot of late shifts from home. This week I'm on that shift again but the setup I have does not agree with the wrist splint I have to wear because of my injury so I've arranged to do the rest of the week in the office.

As of next week the department has annnounced that COVID will no longer be a reason to work from home, so we're all going to be back in the office anyway.
 
arthwollipot said:
As of next week the department has annnounced that COVID will no longer be a reason to work from home, so we're all going to be back in the office anyway.
Click to expand...

Is the pandemic considered sufficiently under control where you are to make that reasonable? I'm in the US Midwest which is currently enjoying a resurgence that dwarfs the previous outbreaks. My employer decided in late summer that we'd continue working from home until June 2021 at the earliest, and that was before things got bad. Unless these vaccines prove to be effective and are distributed in the next couple of months I suspect my employers will push back that June date further.
 
arthwollipot said:
Today is going to be the last time I'll be working from home. During COVID our policy has been that if your shift ends earlier than 4:30 it's done in the office, and if after then, from home. So I've been doing a lot of late shifts from home. This week I'm on that shift again but the setup I have does not agree with the wrist splint I have to wear because of my injury so I've arranged to do the rest of the week in the office.

As of next week the department has annnounced that COVID will no longer be a reason to work from home, so we're all going to be back in the office anyway.
Click to expand...
:confused: We're planning on returning in April.
 
There has been zero community transmission in the ACT for months now. We've had a tiny handful of returning diplomats test positive, who have gone immediately into quarantine. We're small, there's widespread testing, and there is no known wild virus. It's as safe as it can be. Hardly anyone even wears a mask here.
 
The recycling of the databases did not fix the problem, but, have no fear, the technicians at the service providers managed to move the ticket from "new" to "in progress"
 
I appear to have successfully navigated Password Day.

It's a non-trivial exercise for me. I have four passwords that I have to maintain for several different systems, three of which require two-factor authentication using different methods.

At least Password Day only occurs once every three months in this organisation.
 
arthwollipot said:
I appear to have successfully navigated Password Day.

It's a non-trivial exercise for me. I have four passwords that I have to maintain for several different systems, three of which require two-factor authentication using different methods.

At least Password Day only occurs once every three months in this organisation.
Click to expand...

KeePass is your friend.

Sadly, we're moving to Thycotic Secret Server.
 
Status
Not open for further replies.

Back
Top Bottom