Dear Users… (A thread for Sysadmin, Technical Support, and Help Desk people) Part 10

Status
Not open for further replies.
Going to a team Christmas do, but can't shake my confusion about this new program that specifically handles an empty input file, but has a sort step before it that bypasses the program when the input file is empty.

This may come to blows when cold drinks have been taken

Sent from my SM-G973F using Tapatalk
 
Filippo Lippi said:
Similar

Me - "Why are we writing millions of lines of sysout* every day?"

Them - "just in case"


*job output from batch processes
Click to expand...

Heh! I'm like that. I wrote a backup process for my home Linux server that uses rsync to copy data from the primary drive to a backup drive. It runs daily. I compress the output from each run and save it to a file. It's been part of my system since 2014, and I now have about 175 megs in 2,585 bz2 files going back 7 years.

I just did a count: uncompressed, there are 30,368,002 lines total. Most of these files I'll never look at. However, storage is cheap, and I have on occasion looked at the output from a run or two just to see if a certain file was (or was not) included in a backup.

Sometimes it's better to have the data and not use than it is to wish you had it after something goes seriously wrong. In this case, cycling off the output after a few months would probably be a good idea.
 
From the user side: I noticed a few anomalies in some output and was having trouble tracing what had been going on. I asked the sysadmin if activities on that platform were logged, and he said yes. So does anybody review the logs? No. And what happens to the logs? They wait until they've filled the partition and then delete them.

Hmmmm.
 
TragicMonkey said:
Again, someone doesn't understand how numbers work. Yes, it's possible for me to get you all the data you want. However what you want will be between forty and fifty million rows of data. How much time do you want to devote to reading the results? Because I only work 40 hours a week and plan to retire in 2043.
Click to expand...

Heh. Same person I talked out of the above made a new request that would be even bigger. Billions with a B rows of data because she wants three overlapping sets of data where every item on each of the three can (and does) have multiple entries. Pointlessly duplicative--there's no real relation between two of the sets, just that the third connects to both. It makes infinitely more sense to just keep them separate but she has it in her head that "one big spreadsheet" is the way to go. Sure, lady. We'll print it out, it'll cover the entire city and you can drive your car on it to find the rows you want.
 
We have something going on right now that is being referred to as "tender". And it is The Most Important Thing Happening In The Universe Right Now. We literally have to drop everything and do everything they say immediately, because they say so. I have no idea what they're doing. But whatever it is, they're making literally everything else wait.
 
Blue Mountain said:
Heh! I'm like that. I wrote a backup process for my home Linux server that uses rsync to copy data from the primary drive to a backup drive. It runs daily. I compress the output from each run and save it to a file. It's been part of my system since 2014, and I now have about 175 megs in 2,585 bz2 files going back 7 years.

I just did a count: uncompressed, there are 30,368,002 lines total. Most of these files I'll never look at. However, storage is cheap, and I have on occasion looked at the output from a run or two just to see if a certain file was (or was not) included in a backup.

Sometimes it's better to have the data and not use than it is to wish you had it after something goes seriously wrong. In this case, cycling off the output after a few months would probably be a good idea.
Click to expand...
I can get you a great tape library system with probably a few dozen tape cartridges. Should extend your backup capacity to effectively unlimited. Or at least until you or your computer die.
 
TragicMonkey said:
Heh. Same person I talked out of the above made a new request that would be even bigger. Billions with a B rows of data because she wants three overlapping sets of data where every item on each of the three can (and does) have multiple entries. Pointlessly duplicative--there's no real relation between two of the sets, just that the third connects to both. It makes infinitely more sense to just keep them separate but she has it in her head that "one big spreadsheet" is the way to go. Sure, lady. We'll print it out, it'll cover the entire city and you can drive your car on it to find the rows you want.
Click to expand...
Say you will give her an external HDD with the one file on it, and charge her for the cost of the HDD. Because it will be several thousand dollars. Same again, next month. She might...rethink.
 
arthwollipot said:
We have something going on right now that is being referred to as "tender". And it is The Most Important Thing Happening In The Universe Right Now. We literally have to drop everything and do everything they say immediately, because they say so. I have no idea what they're doing. But whatever it is, they're making literally everything else wait.
Click to expand...
I'm a fellow PS sufferer!

Shall I send in some Christmas care packages? A "tender" is not just the process, it's how you feel afterwards.
 
Dear Users: you can't use different logic in different reports to track something not-quite-the-same and then expect the numbers to match. "But Martha's numbers say 71,029!" Yes, because Martha included A, B, C, and excluded D. You excluded A, included D, and for some insane reason decided that Z should be included if and only if it happened during a full moon. That's why your numbers say 40,277 and also you are an idiot. Martha is right, and I for one hope she beats you to death with a taxidermied marmot for questioning her very correct numbers.
 
Shady Chrome websites that disguise their notifications as virus/security pop-ups have been eating up way too much of my time in the last few days.
 
Blue Mountain said:
What's the difference between a Chrome website and a regular website? Is it that the site is intended to give information and or extensions for the Google Chrome web browser?
Click to expand...

Sorry I meant in the sense, from what I've seen at least, it's easier in Chrome to fall victim to these sort of "Let this website send you notifications LOL it's all spam adds that look like virus warnings!" thing.
 
JoeMorgue said:
Sorry I meant in the sense, from what I've seen at least, it's easier in Chrome to fall victim to these sort of "Let this website send you notifications LOL it's all spam adds that look like virus warnings!" thing.
Click to expand...
Chrome used to be a great browser. It isn't any more. I've mentioned before how weird it feels to be saying this, but I find Edge to be more reliable than most other browsers right now.
 
arthwollipot said:
Chrome used to be a great browser. It isn't any more. I've mentioned before how weird it feels to be saying this, but I find Edge to be more reliable than most other browsers right now.
Click to expand...

Yeah Chrome's turn in the "we're the dominant browser so let's bloat it to hell and make it clog everyone's computers" cycle. Firefox did that for a while but their last big rewrite seems to have fixed that. I deliberately don't use Edge (despite hearing good things) as MS's "hold his nose and make him swallow" attitude to deploying edge just pisses me off.
 
It just took three experts 56 minutes to get it into the very thick skulls of management that if particular data isn't put into the database, we cannot pull that data out of the database. I know how much I'm paid per hour, and I'm pretty sure everyone else on the call is paid even more, so this was a waste not just of my time and patience but also money.

How do these people get promoted?!



eta: They've scheduled another call for next week so it can be gone over again, with more people.
 
Last edited:
Status
Not open for further replies.

Back
Top Bottom