| Author |
Message
|
| zpat |
 Posted: Fri Feb 14, 2014 7:18 am Post subject: RFE - vote for IIB web admin to browse the broker error log |
 |
|
Jedi Council
Joined: 19 May 2001
Posts: 5866
Location: UK
|
RFE to provide a means to browse the WMB error log in the IIB web admin interface.
Please vote here
URL: http://www.ibm.com/developerworks/rfe/execute?use_case=viewRfe&CR_ID=45092
_________________
Well, I don't think there is any question about it. It can only be attributable to human error. This sort of thing has cropped up before, and it has always been due to human error. |
|
| Back to top |
|
 |
| Gralgrathor |
| Posted: Fri Feb 14, 2014 7:49 am Post subject: |
|
|
Master
Joined: 23 Jul 2009
Posts: 297
|
Of course if I'm going to have a webinterface for browsing logs, I'd also want to run searches and produce reports. And there are already very kewl tools for collecting and correlating (error) logs available.
But sure, can't hurt to have this.
| Quote: |
| painfully with CAT and GREP commands in the primitive world of the unix command shell |
Seriously? Pounding away at the keyboard grepping my way through TiBs of logging is what makes this job worthwhile! They already took BASIC away from me - don't take this too! |
|
| Back to top |
|
|
| zpat |
| Posted: Fri Feb 14, 2014 8:08 am Post subject: |
|
|
Jedi Council
Joined: 19 May 2001
Posts: 5866
Location: UK
|
No-one's taking anything away. After all Unix is still like it was in 1970's.
I am just trying to get developers to look at the log, which they don't at the moment.
_________________
Well, I don't think there is any question about it. It can only be attributable to human error. This sort of thing has cropped up before, and it has always been due to human error. |
|
| Back to top |
|
|
| Gralgrathor |
| Posted: Fri Feb 14, 2014 8:14 am Post subject: |
|
|
Master
Joined: 23 Jul 2009
Posts: 297
|
| zpat wrote: |
| which they don't at the moment. |
The following will probably tell you something about my skills as a developer: then how dothey develop? Just drag, drop, enter a few lines of code and off to production ? |
|
| Back to top |
|
|
| zpat |
| Posted: Fri Feb 14, 2014 8:37 am Post subject: |
|
|
Jedi Council
Joined: 19 May 2001
Posts: 5866
Location: UK
|
We scrape the error log and generate tickets for them at UAT and PROD.
But it would be nice if they checked for themselves in the early days. Like most things in life - if it's easy people do it....
It's about seeing these log messages earlier in the development cycle, rather than not at all.
_________________
Well, I don't think there is any question about it. It can only be attributable to human error. This sort of thing has cropped up before, and it has always been due to human error. |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Sat Feb 15, 2014 7:52 am Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
I thought a tool like splunk would give that view to them (to the developers) 
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| zpat |
| Posted: Sat Feb 15, 2014 8:10 am Post subject: |
|
|
Jedi Council
Joined: 19 May 2001
Posts: 5866
Location: UK
|
I'm sure it can - but if they can't be bothered to Cat the log ... why would they Splunk it? Integration into the tool makes sense to me.
Hey, one day IBM might let us browse a message without using a third-party tool!
_________________
Well, I don't think there is any question about it. It can only be attributable to human error. This sort of thing has cropped up before, and it has always been due to human error. |
|
| Back to top |
|
|
| Gralgrathor |
| Posted: Sat Feb 15, 2014 9:05 am Post subject: |
|
|
Master
Joined: 23 Jul 2009
Posts: 297
|
| zpat wrote: |
| why would they Splunk it? |
For that matter, why would such a developer bother to open an error-log browser in the webinterface that's undoubtedly going to be as slow as treacle?
Every developer I know already works with a debugger in one window and a tailed error log in another. And more often than not, they don't use the debugger but depend solely on the log.
I'm sure it's a fine idea to include a logview in the web interface, but I think most developers will find that tail and grep are their tools of choice: trigger the flow, and see only the relevant messages scroll by real time. |
|
| Back to top |
|
|
| zpat |
| Posted: Sat Feb 15, 2014 2:18 pm Post subject: |
|
|
Jedi Council
Joined: 19 May 2001
Posts: 5866
Location: UK
|
For goodness sake, I am talking about adding a feature to the GUI, not banning the use of primitive telnet for those antediluvians who like it.
They usually don't logon to the unix host - why should they need to?
_________________
Well, I don't think there is any question about it. It can only be attributable to human error. This sort of thing has cropped up before, and it has always been due to human error. |
|
| Back to top |
|
|
| Gralgrathor |
| Posted: Sat Feb 15, 2014 2:53 pm Post subject: |
|
|
Master
Joined: 23 Jul 2009
Posts: 297
|
| zpat wrote: |
| For goodness sake |
Don't fret; I'm not trying to talk anybody out of anything.
| zpat wrote: |
| They usually don't logon to the unix host |
Well, every developer I know does. How else are you going to see if the code you just deployed doesn't choke on production(-like) data? You write code, deploy it to the DEV environment, run some data against it, note problems in the log, and then use the debugger to pinpoint the problem, if it's that kind of problem. You're not going to deploy anything to UAT that hasn't been technically tested against a load of production-like data, and you're not going to run that kind of test through a debugger. So anybody who's not just playing around with this stuff *will* have a log tailed while they're triggering the flow. And that means logging on to the unix host. Sysadmins usually have protocols for granting developers sufficient rights on the DEV (and sometimes TEST) environment for viewing logs, and often enough grant them additional rights so that they can kill their own infinite loops 
I think it's a good idea to have some webinterface for viewing errors - but not for developers: for system- and application managers. And they wouldn't need to view all of the log: they'd need to see a summary that identifies problems during production runs, so that they can notify the right parties to get to work on the issue.
I like splunk, except for the fact that it costs way too much. |
|
| Back to top |
|
|
| mqjeff |
| Posted: Sun Feb 16, 2014 8:25 am Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
| Please do try to remember that the set "every developer" is not the same as the set "every developer I know". |
|
| Back to top |
|
|
| zpat |
| Posted: Sun Feb 16, 2014 9:24 am Post subject: |
|
|
Jedi Council
Joined: 19 May 2001
Posts: 5866
Location: UK
|
Support pac MS0P allows you to browse MQ logs from MQ explorer - it's a convenience.
People do things more often when they are convenient. Anyone living in the land of drive through banks should know that ...
_________________
Well, I don't think there is any question about it. It can only be attributable to human error. This sort of thing has cropped up before, and it has always been due to human error. |
|
| Back to top |
|
|
| Gralgrathor |
| Posted: Mon Feb 17, 2014 2:09 pm Post subject: |
|
|
Master
Joined: 23 Jul 2009
Posts: 297
|
| mqjeff wrote: |
| Please do try to remember that the set "every developer" is not the same as the set "every developer I know". |
Which is why I specifically say "every developer I know". I am well aware that there may be oddballs out there who manage to develop and deploy to UAT something without actually testing it against a load of data. I just can't imagine what such a person would look like.
Well, probably not much worse than me though. |
|
| Back to top |
|
|
| mqjeff |
| Posted: Mon Feb 17, 2014 2:16 pm Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
| Gralgrathor wrote: |
| mqjeff wrote: |
| Please do try to remember that the set "every developer" is not the same as the set "every developer I know". |
Which is why I specifically say "every developer I know". I am well aware that there may be oddballs out there who manage to develop and deploy to UAT something without actually testing it against a load of data. I just can't imagine what such a person would look like.
Well, probably not much worse than me though. |
I was suggesting, ever so gently, that "every developer you know" is a very elite set of well-trained, experienced, sensible people.
Which is not remotely the same as the set that a lot of us usually work with. |
|
| Back to top |
|
|
| Gralgrathor |
| Posted: Mon Feb 17, 2014 2:34 pm Post subject: |
|
|
Master
Joined: 23 Jul 2009
Posts: 297
|
| mqjeff wrote: |
| I was suggesting, ever so gently, that "every developer you know" is a very elite set |
No they're not. They're a bunch of wankers, no better'n me. But they still understand you can't stick your hand in a jar without taking the lid off. |
|
| Back to top |
|
|
|
|
