Recently there have been a couple of instances of spam I would have sworn were the same user. Prior to the d.o upgrade, I could peruse the log files and find the IP addresses of the users in question. It was time consuming, but at least doable.
The logs link is now gone from the admin page. I don't know if it was an oversight or intentional, but would it be possible to have the user's last used IP address visible on their account page somehow? Perhaps with a block ip link? The spammers that create multiple accounts irritate me more than any others-- it would be nice to be able to identify and block them easily.
Comments
Comment #1
WorldFallz commentedI just spent the last hour removing major link spam:
http://drupal.org/node/424262
http://drupal.org/node/424268
http://drupal.org/node/424270
http://drupal.org/node/424274
http://drupal.org/node/424276
http://drupal.org/node/424250
http://drupal.org/node/424278
That's one less hour I could have spent assisting users in the forums. =/
Judging by the posts, I would guess they were from the same IP or IP range. We need to add some sort of easier anti-spam measures for maintainers. I'm not exactly sure what that is though. Perhaps a "block ip" and 'deleted all posts/comments from this user" on user pages. I realize it would have to be used carefully, so as not to delete legitimate downstream posts, but in this case there were NO downstream posts-- it was page after page of pure linkspam. Hopefully someone with better first hand knowledge has some ideas.
Clearly the spammers have increased their targeting of d.o-- we need to figure out something so maintainers don't waste all their time handling spam instead of helping users.
any one have any ideas?
Comment #2
WorldFallz commentedAnd 2 more of the same style users:
http://drupal.org/user/475926
http://drupal.org/user/475928
Comment #3
WorldFallz commentedAnd yet another example:
http://drupal.org/node/426038
We're being overrun with spammers lately-- there has to be something we can do to make this easier for maintainers to handle.
Since we require registration for posting-- maybe a registration captcha? And a simple way for maintainers to block an ip from the user's account page?
Comment #4
gerhard killesreiter commentedthe flag module will be deployed soonish and should help with this.
Comment #5
sivaji_ganesh_jojodae commentedPerhaps mollom could help us better.
Comment #6
gregglesI think mollom is a relatively good idea, but not perfect.
Fasttoggle would also make the job of blocking a user faster.
Comment #7
sivaji_ganesh_jojodae commentedFasttoggle sounds good but my interest is to make the spam posts removal automated what you suggest is to ease the administrative task. I can't comment on mollom's performance because I haven't used it for any large sites. From what I know mollom is always improving - it will be a mutual win if d.o opts for mollom. I have always wondered why that didn't happen as yet. Perhaps a combination of fasttoggle and mollom can make things much better.
Comment #8
sunMollom would be more than happy to help.
The issue has been discussed in the past already. (At least last time I heard about it) the intention was to limit it to new users only; i.e., limited to the first 24-48 hours, first X amount of posts, or whatever. That should be possible to do now, as I also just mentioned in the corresponding Mollom issue #706180: Fire a hook before Mollom adds its stuff in hook_form_alter()
Comment #9
greggles@sivaji - currently removing spam comments requires 7 clicks. Installing fasttoggle would remove 3 of them and add 1 so a net reduction of 2.
1. Click edit
2. Click blocked
3. Click save
4. Click Administer comments
5. Click checkbox for all comments
6. Click delete
7. Click confirm
Fasttoggle removes 1-3 but adds a new one: Click fasttoggle link to block user. It also drastically reduces the scrolling required for step 2/3.
Comment #10
sivaji_ganesh_jojodae commentedI understand that. In other words, it needs a human moderator to do this. Let's free him and mollom (or equivalent spam control program) do that job.
Comment #11
naught101 commentedIf we used flag, (which will probably be rolled out within the next few days), then we could have a "mark as spam" flag, as well as a "mark as redundant" flag, which might help clear up old subscribe comments as well (See #1303644: comment validation to prevent '+1 subscribe' comments)
There is also the spam module, which needs a lot of love, but is potentially useful here.
Comment #12
silverwing commentedThis was recently brought up here: #1293186: Spam - meta: better spam-combating suggestions
I suggest starting by installing http://drupal.org/project/hidden_captcha for spambot registrations
Flag would work, but I believe it works best for spam with flag_abuse. Then we work towards getting spam.module ready for D7
Comment #13
MGParisi commentedDelay email notifications. I have a feeling spammers dont care if you delete their posts because they know Drupal just sent out their spam to everyone on the mailing list. If we make it less likely that their spam will get to email then they will be less likely to post spam.
Comment #14
sunYou can be sure that spammers/spambots don't know any specifics about Drupal or drupal.org.
Also note that Flag module will be deployed very soon, but it will not be used for comments due to performance issues.
Comment #15
naught101 commentedmarked #226678: Add a "Report spam/abuse" link to forum/issue comments (next to the "edit" & "reply" links). as a duplicate.
Might be worth looking through https://drupal.org/project/issues/search?text=spam&projects=Drupal.org+w... for more...
Comment #16
klonosPlease also take a look at my proposal here: #1308176: [meta] Battle plan for stopping spam/"subscribe"/"+1"/"thank you" comments (and cleaning up old ones from the db too).. Thanx in advance.
Comment #17
sk33lz commented@sun That's funny because it is already being done... I received this in my inbox as I am a member of the Consulting and Business group, so I don't believe you can say that the spammers have no idea what they are doing on Drupal.org web properties.
Please take a look here, http://groups.drupal.org/node/182814.
This qualifies as an event? http://groups.drupal.org/node/182549
Both of those posts were posted within 24 hours of each other and exactly 35 minutes apart. They were both posted by a user of about 2 weeks, http://groups.drupal.org/user/1192469, which links to their website in their profile and has no other information aside from a name really. In fact, that website is not even a Drupal website. It is a Wordpress website, which has no content and links to a Wordpress theme.
Comment #18
sk33lz commented+1 for Mollom integration.
I am not sure the performance implications for such an install on Drupal.org, but it works great for all of the sites I have used it on so far. I used to spend hours each week dealing with spam comments, but now I don't even have to think twice about them.
It would at least do 1/2 the work of getting rid of the obvious spam. That is better than nothing when we are talking about volunteers and users doing moderating.
Hopefully Dries doesn't charge Drupal.org for usage :)
Comment #19
dddave commentedI want to point out that Mollom can be obtrusive as any CAPTCHA device. Especially when non-native writers make spelling errors or such. I don't know what Mollom can do now to avoid hassling legitimate users but I suspect not too much.
Perhaps if there is an treshold (e.g. only for users who are on d-org for less than a month; for users with less than 50 comments or something like this) and only new users get hassled...
Comment #20
danepowell commentedI agree with #19- I'm sorry I don't have time to form a full and cogent argument, but I am strongly against using Mollom on Drupal.org, unless it can be configured EXTREMELY carefully to be almost 100% transparent to legitimate users. The workload and frustration that I have experienced as an end user trying to appease Mollom, imposed on thousands of legitimate users, would not nearly justify using it to prevent a few SPAM posts here and there.
Comment #21
klonosYeah, I too believe that we should not be bothering legitimate (long-time) members with CAPTCHA-like steps when they try to post comments. I agree that if such a measure is taken it should be implemented in a way that will only be in effect for new accounts with only a handful of comments/edits.
Comment #22
michelleChanging tag to unify these under one tag.
Comment #23
cweagansI know this is an older issue, but I'm going to mark this as a duplicate of #1293186: Spam - meta: better spam-combating suggestions, as the other issue has more up to date, actionable information in it.