NUTCH-2573 Suspend crawling if robots.txt fails to fetch with 5xx status #724
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- add properties
http.robots.503.defer.visits :
enable/disable the feature (default: enabled)
http.robots.503.defer.visits.delay :
delay to wait before the next trial to fetch the deferred URL
and the corresponding robots.txt
(default: wait 5 minutes)
http.robots.503.defer.visits.retries :
max. number of retries before giving up and dropping all URLs from
the given host / queue
(default: give up after the 3rd retry, ie. after 4 attempts)
- handle HTTP 5xx in robots.txt parser
- handle delay, retries and dropping queues in Fetcher
Stop adding fetch items if timelimit is reached
- redirects (http.redirect.max > 0) or
- outlinks (fetcher.follow.outlinks.depth > 0)
lewismc
requested changes
Jan 15, 2022
There was a problem hiding this comment.
I like this patch. It's actually something which will help us loop back in with website administrators to notify them of service issues. Thanks for including the metrics this is really useful. Some minor suggestions from me @sebastian-nagel .
| @@ -263,6 +283,10 @@ public synchronized int checkExceptionThreshold(String queueid) { | |||
| return 0; | |||
| } | |||
|
|
|||
| public int checkExceptionThreshold(String queueid) { | |||
- rename counter to follow naming scheme of other robots.txt related counters: `robots_defer_visits_dropped` - rename method timelimitReached -> timelimitExceeded - add Javadoc
|
Hi @lewismc, done: updated metrics wiki page (hitByTimeLimit is already documented), added Javadocs and renamed the counter to follow the naming convention of the other robots_* counters. Also renamed the method ("timelimitReached" -> "timelimitExceeded"). |
|
Looks like it failed on Javadoc generation @sebastian-nagel |
lewismc
approved these changes
Jan 17, 2022
- fix javadoc error
sebastian-nagel
force-pushed
the
NUTCH-2573-suspend-crawling-robotstxt-fails
branch
from
170e3fc to
7f944c1
Compare
sebastian-nagel
deleted the
NUTCH-2573-suspend-crawling-robotstxt-fails
branch
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
http.robots.503.defer.visits:enable/disable the feature (default: enabled)
http.robots.503.defer.visits.delay:delay to wait before the next trial to fetch the properties
(default: wait 5 minutes)
http.robots.503.defer.visits.retries:max. number of retries before giving up and dropping all URLs from the given host / queue
(default: give up after the 3rd retry, ie. after 4 attempts)
Stop queuing fetch items if timelimit is reached
In a first version, I forgot to verify whether the Fetcher timelimit (
fetcher.timelimit.mins) was already reached before re-queuing the fetch item. This caused very few fetcher task to end up in an infinite loop. In detail, this happened:Then steps 1 and 3 are retried until the max number of retries is reached. But this was fixed and I've also made sure that redirects or outlinks are not queued if the timelimit is reached.