Better differentiation between exceptions and logs

[priestc]

At my job, I'm in charge of error reporting and logging. During our last release, I went through and changed a bunch of:

try:
    [code]
except:
    pass

to

try:
    [code]
except Exception as exc:
    log.error(exc)

The problem is that our sentry "messages/day" went from around 80/day to about 350/day. All day people were asking me why the site was blowing up. I had to explain to them that logs and exceptions are two different things. The exception rate is the same, logger rate went up.

It would be nice if sentry made more of an effort to differentiate between logger statements and exceptions. Maybe color code the orange circle? Red for exceptions, blue for info, green for debug, orange for warning, etc...

[dcramer]

The way logging treats exceptions are generally as "error" level. You can
send other levels, and they will be shown differently in the UI. For

example logging.warning is orangish.

David Cramer
http://disqus.com/zeeg
http://twitter.com/zeeg

On Sun, Apr 22, 2012 at 7:12 AM, priestc <
reply@reply.github.com

wrote:

At my job, I',m in charge of error reporting and logging. During our last
release, I went through and changed a bunch of:

try:
[code]
except:
pass

to

try:
[code]
except Exception as exc:
log.error(exc)

The problem is that our sentry "messages/day" went from around 80/day to
about 350/day. All day people were asking me why the site was blowing up. I
had to explain to them that logs and exceptions are two different things.
The exception rate is the same, logger rate is something different.

It would be nice if sentry made more of an effort to differentiate between
logger statements and exceptions. Maybe color code the orange circle? Red
for exceptions, blue for info, green for debug, orange for warning, etc...

---

Reply to this email directly or view it on GitHub:
https://github.com/dcramer/sentry/issues/449

[heynemann]

@priestc Can you please close this as it's not an issue?