Read just our best, feature-length articles without all the extra stuff.
They Only Had a Small Amount of Error Reports Queued Up [Humorous Image]
Enter your email below to get exclusive access to our best articles and tips before everybody else.
This is a situation where you are probably better off not knowing the details…
I’m sure this is some sort of record [Imgur]
Akemi Iwaya (Asian Angel) is our very own Firefox Fangirl who enjoys working with multiple browsers and loves 'old school' role-playing games. Visit her on Twitter and Google+.
Looks pretty NORMAL to me.
A few months ago, an email server I run was constantly running out of storage even though it has 4x the amount of storage that all the user quotas combined would require. I cleared out the log files and it helped a little, but not enough. Every week or so, it would fill up again. I’d get busy and forget about it, then email would start bouncing when it was 100% full. Unhappy users would call. Not good.
Jump forward 2 months and I’m doing a systems upgrade – both the version of the email system AND OS are being upgraded. The first thing I do after all the planning is complete is a full system backup. Everything. Previously, this process was taking about 30 minutes, but since an automatic backup happened just a few hours earlier, I expected the backup to complete in 3-4 minutes. It took the 30+ minutes, so something was definitely wrong. The only way that could happen is if there were thousands of files that changed between 2am-4am OR 1 really huge file that changed.
Turned out it was a single HUGE Log file for a java application not stored in the expected place for log files, 11GB worth of the same error being appended every 20 seconds. It had been for … perhaps 2 years. I love Java, NOT! Every recent backup was 11GB larger than it needed to be; compression made it about 400M on disk, but still, it was big. The log size was probably growing about 500MB per month. Googled the error, found a solution, applied it and that file has remained 0 bytes for a few months. The OS and system updates went according to plan.
At least it wasn’t 35.2GB log file.
Thanks for info. It helped me a lot in the same problem i was facing in my Project :)
It would be interesting to see inside the log file(s). Probably nothing but banal repititions there, but might point to something that needs to be addressed.
I fail to see how it’s Java’s fault in that situation TheFu. This is more like an incompetent programmer’s fault.
It’s a combination of things.
1. No one checking the logs – and then clearing them.
2. Bad logging design – frequently piles of useless info but ever enough on an actual error.
3. Bad programming, worse debugging.
4. No viable routine for determining where disk space is being wasted.
A meter was originally defined as one ten-millionth the distance between the Earth’s equator and the North Pole.
Enter your email address to get our daily newsletter.
Enter Your Email Here to Get Access for Free: