logstash stops processing when an anomaly in event stream is found

Dont know if this question/bug is

I am ingesting the AWS ELB access logs and at times [due to how ELB logs]; there are entries like this:

```
2015-05-15T21:51:49.335698Z elb-production 31.3.226.2:38778 10.0.55.56:80 0.000041 0.00055 0.000071 301 301 0 184 "GET http://55.48.116.183^M^M
Accept:80/user.php?action=users-Login HTTP/1.1"
```

The above in principle must be a single line and without the carriage [^M^M], but that is how ELB is reporting it and there is nothing we can do about it I guess. But the moment this is encountered, logstash stops processing with an exception.

So my question is
- Is there a grok which can catch this type of anomaly in a otherwise large stream of data.
- If not how can we skip such events and get to the next event. Ideally it would be nice, if logstash would go on to next event.