KOHA-MATTERMOST:~$ free -hKOHA-MATTERMOST:~$ uptimeKOHA-MATTERMOST:~$ df -h← Previous day | Today | Next day → | Search | Index
All times shown according to UTC.
| Time | Nick | Message |
|---|---|---|
| 05:05 | thd | dcook: What does one need to do for bugzilla to ban an account? |
| 05:17 | dcook | Have too many URLs in a comment I think |
| 05:17 | Possibly also too many comments too quickly | |
| 05:17 | But the URL thing seems to be the key factor | |
| 05:58 | reiveune joined #koha | |
| 05:58 | reiveune | hello |
| 06:11 | magnuse | \o/ |
| 06:51 | cait joined #koha | |
| 06:53 | cait | good morning #koha! |
| 07:07 | matts | hi koha! |
| 07:07 | paul_p joined #koha | |
| 07:09 | thibaud_g joined #koha | |
| 07:11 | matterbridge | [mattermost] <paulderscheid> morning #koha |
| 07:11 | cait | paulderscheid[m]: did I miss the instructions on how to get on mattermost? |
| 07:11 | bonjour matts :) | |
| 07:12 | Roles page is up: https://wiki.koha-community.or[…]i/Roles_for_24.11 | |
| 07:21 | paulderscheid[m] | seems so cait |
| 07:21 | :D | |
| 07:21 | matterbridge | [mattermost] <aulda> We're just in Beta at the moment... |
| 07:22 | ashimema joined #koha | |
| 07:22 | ashimema | it's close.. very close 😉 |
| 07:22 | we're just awaiting dns stuff to migrate through... ssl etc | |
| 07:22 | paulderscheid[m] | I love it |
| 07:22 | matterbridge | [mattermost] <aulda> Should have the correct URL sorted today so that we can launch more formally |
| 07:24 | [mattermost] <aulda> Also, I saw the call for the community newsletter so I'll draft a piece for that, and for the mailing lists giving people heads up about Mattermost. | |
| 07:24 | [mattermost] <paulderscheid> That's a very good idea! | |
| 07:25 | cait | aulda: not sure if you can see that, but sounds good, just let me know when we are supposed to sign up |
| 07:25 | I think the URL shoudl be sorted from what I saw (email between rangi and tcohen) | |
| 07:25 | matterbridge | [mattermost] <paulderscheid> Yep, the bridge is bi-directional :) |
| 07:25 | [mattermost] <aulda> I see you @cait | |
| 07:25 | [mattermost] <aulda> 😉 | |
| 07:41 | cait | aulda: matterbridge on the agenda for Wednesday? :) I just added some other topics as well |
| 07:41 | https://wiki.koha-community.or[…]ing_17_April_2024 | |
| 07:45 | paulderscheid[m] | Beta is over :D |
| 08:06 | ashimema | Yeay.. address is now live proper by the looks |
| 08:06 | chat.koha-community.org | |
| 08:17 | interesting | |
| 08:19 | did the bridge stop already | |
| 08:28 | magnuse | oh noes |
| 08:34 | signing up for mattermost was pretty smooth, at least | |
| 08:40 | matterbridge joined #koha | |
| 08:40 | matterbridge joined #koha | |
| 08:41 | matterbridge joined #koha | |
| 08:41 | matterbridge joined #koha | |
| 08:42 | matterbridge joined #koha | |
| 08:42 | matterbridge joined #koha | |
| 08:42 | matterbridge joined #koha | |
| 08:43 | matterbridge joined #koha | |
| 08:43 | matterbridge joined #koha | |
| 08:44 | perplexedtheta joined #koha | |
| 08:44 | matterbridge | [mattermost] <PerplexedTheta> The bridge went down because I didn't update the config with the new URL - it should be back up now |
| 08:44 | [mattermost] <PerplexedTheta> chat.koha-community.org is live! | |
| 08:52 | Annelisterman[m] joined #koha | |
| 08:52 | Annelisterman[m] | What is this mattermost? Are we supposed to join it instead of this IRC? |
| 08:53 | matterbridge | [mattermost] <aulda> At Hackfest we discussed a new way of enabling community communication and chat, and settled on Mattermost as the best option for us. |
| 08:54 | [mattermost] <aulda> We have been setting up a Mattermost server (hosted by PTFS Europe), and testing the functionality in the last few days. | |
| 08:54 | thd | Annelisterman[m]: I believe that there is intended to be a bridge, there it is. |
| 08:54 | matterbridge | [mattermost] <aulda> I am going to pop an announcement in the Koha Community newsletter and the maillists in the coming days. |
| 08:55 | [mattermost] <aulda> Yes, the bridge between Mattermost and IRC enables those on Mattermost to see the IRC chat and vice-versa. | |
| 08:56 | Annelisterman[m] | Ok. Is this IRC going to stay active though? Or maybe you are explaining it all in the newsletter. :) |
| 08:58 | thd | Annelisterman [m]: IRC will be active as long as people want it to be by participating there and to whatever extent the chat/IRC bridge works. |
| 08:58 | matterbridge | [mattermost] <aulda> I will explain the discussion that we had at Hackfest in the newsletter, yes. The intention, for now, is that IRC stays active. It will then depend on how active it is in the long term I would think, and how inclusive Mattermost proves to be. |
| 09:00 | Annelisterman[m] | Ok. We'll have to look into it. We are now using IRC with Matrix/Element at the moment. |
| 09:00 | s/now// | |
| 09:03 | matterbridge | [mattermost] <PerplexedTheta> Now enabled IRC Bridge as an auto-joined channel, so nobody misses out on the fun |
| 09:20 | magnuse_ joined #koha | |
| 09:24 | cait | Annelisterman[m]: ultimately the idea will be to use a newer tool, that also allows channels etc. and gather the community there |
| 09:25 | but we are testing for now, IRC is not going away, but it will not be able to offer the same functionality, for example people posting pictures, different channels, settng your own status as a new solution will provide | |
| 10:11 | tcohen | cait wb to work :-D |
| 10:11 | bug 36593 for you | |
| 10:11 | huginn` | 04Bug https://bugs.koha-community.or[…]_bug.cgi?id=36593 enhancement, P5 - low, ---, tomascohen, Needs Signoff , Add support for the `time` column type on TestBuilder |
| 10:23 | matterbridge | [mattermost] <cait> tcohen++ thank you! Will check a little later today, still working on the backlog of mails |
| 10:23 | [mattermost] <cait> but RM time is planned for too :) | |
| 10:23 | thd | tcohen: I thought that you may have been in France. |
| 10:25 | cait | we have been :) |
| 10:26 | thd | cait: I saw tcohen on IRC at an earlier than usual hour. |
| 10:31 | tcohen | hi thd |
| 10:31 | jetlag | |
| 10:31 | I woke at 5 | |
| 10:34 | cait | ouch |
| 10:40 | thd | I have been up but my sleep schedule was somewhat rearranged as struggled to find my typo. I left a letter out of my username and never noticed while rechecking my password. |
| 10:42 | tcohen: When would you have have time later today or tomorrow to pre-setup docker-mailserver? | |
| 10:44 | tcohen: Fixing that: Would you have time later today or tomorrow to pre-setup docker-mailserver? | |
| 10:55 | oleonard joined #koha | |
| 10:56 | oleonard | Hi #koha |
| 10:56 | Sorry to any who are back at work already with a Hackfest Hangover | |
| 10:56 | thd | tcohen: I am writing rangi about updating DNS now. He will probably be asleep. |
| 10:59 | tcohen | thd I don't think I'll have time today or tomorrow, catching up with jetlag and a pile of emails |
| 10:59 | thd | tcohen: Were you also in France? |
| 11:00 | tcohen | can you drop me an email about what you need to do, and I try to think about it in between things? |
| 11:00 | thd yeah | |
| 11:01 | thd | Something like what we did to add the wiki container to portainer. In this case, there is no database to worry about. |
| 11:01 | tcohen | so you want to set a postfix server on a container? |
| 11:02 | thd | It is a little more than Postfix but Postfix is essentially all we would be using. |
| 11:03 | I have been testing the configs and I have a very good configuration after breaking things a few times. | |
| 11:04 | Dovecot configuration can be improved but we only presently use Dovecot for authentication and transfering bounces somewhere else as needed. | |
| 11:07 | tcohen: The big accomplishment was a clever trick a manged to get working where some addresses could send mail directly to any server without being blocked by the necessary configuration of relay functions. | |
| 11:08 | Everything is sender dependent so that you can have different functions for different senders without one function clobbering everything. | |
| 11:09 | ... and it delivers to Gmail, etc. mailboxes via secure-mailgate.com passing all tests. | |
| 11:20 | thibaud_g joined #koha | |
| 11:21 | tcohen | thd I like where you're headed to, but in Marseille there was a comment |
| 11:21 | about finding a better SMTP provider that does DKIM | |
| 11:22 | the one HKS3 had found was not good enough, and they are committed to finding another one | |
| 11:23 | thd | Yes, I know that option is being explored, but that will add more proprietary services which a FOSS project should not depend upon especially when these services are prone to suspend and then cancel accounts over temporary rises in bounced mail. |
| 11:24 | Is HKS3 themselves not satisfied with secure-mailgate.com? | |
| 11:24 | tcohen | got it |
| 11:24 | they only do DKIM if you handle your domain with them I think | |
| 11:25 | that's what I understood | |
| 11:28 | thd | Discourse officially recommends these proprietary SMTP services as the easy way to do email but that is because many adopters find setting up a mailserver too challenging. However, every one of these services has unhappy users who briefly had too many spammers trying to register Discourse accounts with someone else's email address etc. which then produces bounces as one way of getting too many bounces. |
| 11:31 | Mail relays are easy. The tricky thing is configuring so that everything can work and some senders on the system can bypass the relay configuration. | |
| 11:32 | Or if service has been suspended because some problem is unavoidable a minor change in configuration and restarting the container has delivery working again without a new proprietary service. | |
| 11:32 | tcohen | thd, conceptually I understand all you're talking about. |
| 11:33 | Please let me know on an email (so it doesn't get lost for me) how I can help you | |
| 11:33 | and I'll start thinking about it in background and come back to you once my brain is back to a normal-ish place | |
| 11:33 | timezona | |
| 11:35 | thd | The first email you will have is where you and paulderscheid[m] will be on the CC in the message to rangi about DNS changes for anything to work including any new proprietary system. |
| 11:38 | thibaud_glt joined #koha | |
| 11:38 | thd | Also, if you have enough brain power at the moment, I am not certain where to ask rangi to direct DNS for an additional testing instance of docker-mailserver. An additional instance we can break without disrupting things where port 25 is not on the same IP address. |
| 11:43 | tcohen: Aside from services such http and https I think Traefik has trouble managing things other than DNS certificates. Email protocols are sometimes a problem for Traefik if Traefik is put in the middle for multiple mail services on the same port. | |
| 12:58 | tcohen | thd I never used traefik like that. I know it does SNI and HostSNI can be used to match things, but I never used it really |
| 13:11 | thibaud_glt joined #koha | |
| 13:17 | thd | tcohen: Assuming that Traefik may be a problem when acting as a proxy for email protocols but can be bypassed for mail server ports. Is there an additional IP address on that machine or anther system where we can run a test instance of docker-mailserver such that a production instance and a test instance do not conflict for running the same services with the same email standards specified ports? |
| 13:17 | RayVoelker[m] left #koha | |
| 13:27 | cait1 joined #koha | |
| 13:49 | khall joined #koha | |
| 13:54 | thibaud_g joined #koha | |
| 13:56 | caroline joined #koha | |
| 14:16 | cait1 joined #koha | |
| 14:57 | thibaud_glt joined #koha | |
| 15:00 | thibaud_glt1 joined #koha | |
| 15:20 | bag joined #koha | |
| 15:32 | matterbridge | [mattermost] <aulda> **Mattermost - the new place to be for all things Koha** |
| 15:32 | [mattermost] <aulda> [chat.koha-community.org](http://chat.koha-community.org) | |
| 15:32 | [mattermost] <aulda> Many of us have been discussing for ages how we would like to improve communication across the Koha Community and these discussions really came to a head at Hackfest last week in Marseille. | |
| 15:32 | [mattermost] <aulda> We all decided that we needed a new tool that is as inclusive as possible, open source, modern, easy to use and familiar. The tool selected is [Mattermost](https://mattermost.com/) and you can start to use it straight away to ask questions, discuss bugs, raise issues or find out news about the Koha project. | |
| 15:32 | [mattermost] <aulda> It will take a few weeks to settle into, and to implement all of the connections, integrations and bots that will make it as near perfect as a tool can be, but if you are happy to just dive in, then go over to [chat.koha-community.org](http://chat.koha-community.org) and log in. | |
| 15:32 | [mattermost] <aulda> There is currently a bridge to IRC so that anything posted in IRC appears in a channel in Mattermost and anything posted in that channel (in Mattermost) appears in IRC. There are no plans to leave IRC completely - it will take time for everyone to get used to Mattermost and for the community to make sure it isn’t cutting out any voices if it does decide to stop using IRC. | |
| 15:32 | [mattermost] <aulda> The current Mailing lists are also unaffected by this change. | |
| 15:32 | [mattermost] <aulda> | |
| 15:32 | [mattermost] <aulda> PTFS Europe is hosting the Koha Community Mattermost server and there are a number of admin users across the community. | |
| 15:43 | [mattermost] <matts> Hi from mattermost :) | |
| 15:47 | Topic for #koha is now Welcome to #koha this channel is for discussion of the Koha project and software https://koha-community.org - we have also opened up a mattermost server where most people in the koha-community have joined for discussion on all things Koha - chat.koha-community.org | |
| 15:47 | Topic for #koha is now Welcome to #koha this channel is for discussion of the Koha project and software https://koha-community.org - we have also opened up a mattermost server where most people in the koha-community have joined for discussion on all things Koha - https://chat.koha-community.org | |
| 15:47 | domm[m] joined #koha | |
| 15:47 | domm[m] | I get a blank page on firefox, and "Limit exceeded" via google-chrome - I assume some finetuning is still needed :-) |
| 15:48 | bag | so do I |
| 15:48 | hmm | |
| 15:48 | domm[m]: try this https://chat.koha-community.org/landing#/ | |
| 15:49 | matterbridge | [mattermost] <jpahd> oh no |
| 15:50 | [mattermost] <PerplexedTheta> I am not surprised. The server currently rate-limits at 10 queries per second, thats probably not enough now we have 35 people! | |
| 15:50 | [mattermost] <jpahd> Ah | |
| 15:50 | [mattermost] <jpahd> Ok | |
| 15:50 | [mattermost] <PerplexedTheta> Let me up this and quickly restart the server | |
| 15:50 | domm[m] | JS not loading |
| 15:50 | EvalError: call to eval() blocked by CSP | |
| 15:51 | matterbridge | [mattermost] <jpahd> As long as we don't need a k8s cluster to run our chat :D |
| 15:52 | [mattermost] <PerplexedTheta> ```bash | |
| 15:52 | [mattermost] <PerplexedTheta> ptfsKOHA-MATTERMOST:~$ free -h |
|
| 15:52 | [mattermost] <PerplexedTheta> total used free shared buff/cache available | |
| 15:52 | [mattermost] <PerplexedTheta> Mem: 1.9Gi 1.1Gi 73Mi 41Mi 1.0Gi 850Mi | |
| 15:52 | [mattermost] <PerplexedTheta> Swap: 2.0Gi 81Mi 1.9Gi | |
| 15:52 | [mattermost] <PerplexedTheta> ptfsKOHA-MATTERMOST:~$ uptime |
|
| 15:52 | [mattermost] <PerplexedTheta> 15:52:10 up 4 days, 1:31, 1 user, load average: 0.36, 0.28, 0.22 | |
| 15:52 | [mattermost] <PerplexedTheta> ptfsKOHA-MATTERMOST:~$ df -h |
|
| 15:52 | [mattermost] <PerplexedTheta> Filesystem Size Used Avail Use% Mounted on | |
| 15:52 | [mattermost] <PerplexedTheta> udev 965M 0 965M 0% /dev | |
| 15:52 | [mattermost] <PerplexedTheta> tmpfs 197M 848K 196M 1% /run | |
| 15:52 | [mattermost] <PerplexedTheta> /dev/vda1 59G 17G 41G 30% / | |
| 15:52 | [mattermost] <PerplexedTheta> tmpfs 984M 1.1M 983M 1% /dev/shm | |
| 15:52 | [mattermost] <PerplexedTheta> tmpfs 5.0M 0 5.0M 0% /run/lock | |
| 15:52 | [mattermost] <PerplexedTheta> /dev/vda15 124M 12M 113M 10% /boot/efi | |
| 15:52 | [mattermost] <PerplexedTheta> /dev/sda 20G 312M 19G 2% /data/compose/55/volumes/app/mattermost/data | |
| 15:52 | [mattermost] <PerplexedTheta> tmpfs 197M 0 197M 0% /run/user/1000 | |
| 15:52 | [mattermost] <PerplexedTheta> ``` | |
| 15:52 | [mattermost] <PerplexedTheta> We have lots of room to grow 😉 | |
| 16:05 | dpk joined #koha | |
| 16:08 | dpk | I have a question for the experts here.. We are seeing loads of these errors: |
| 16:08 | [2024/04/15 17:07:46] [WARN] No method count found for Koha::Items DBIx::Class::SQLMaker::ClassicExtensions::puke(): Fatal: NULL-within-IN not implemented: The upcoming SQL::Abstract::Classic 2.0 will emit the logically correct SQL instead of raising this exception. at /usr/share/koha/lib/Koha/Objects.pm line 583 | |
| 16:08 | at /usr/share/koha/lib/C4/Search.pm line 2049. | |
| 16:09 | And similar error at a lesser rate for some other queries like /usr/share/koha/opac/cgi-bin/opac/opac-ISBDdetail.pl line 188. | |
| 16:11 | This has been exacerbated today because the Amazon bots descended on us. I found that they were using a URL of /opac/opac-search.pl instead of /cgi-bin/opac/opac-search.pl so the recommended robots.txt file was not effective. | |
| 16:11 | I have updated robots.txt, but it will take hours before they re-read it and stop. | |
| 16:11 | matterbridge | [mattermost] <ashimema> @perplexedtheta isn't this what we saw today? |
| 16:11 | dpk | Still interested in this NULL-within-IN error though. |
| 16:13 | matterbridge | [mattermost] <PerplexedTheta> Yeah it is |
| 16:13 | [mattermost] <mblenk> [Bug 29087](https://bugs.koha-community.or[…]bug.cgi?id=29087) | |
| 16:13 | huginn` | 04Bug https://bugs.koha-community.or[…]_bug.cgi?id=29087 major, P5 - low, ---, koha-bugs, NEW , Holds to pull list can crash with a SQL::Abstract puke |
| 16:13 | 04Bug 29087: major, P5 - low, ---, koha-bugs, NEW , Holds to pull list can crash with a SQL::Abstract puke | |
| 16:14 | matterbridge | [mattermost] <mblenk> Although it says 'Holds to pull' it uses the same method |
| 16:14 | [mattermost] <PerplexedTheta> Ha damn, thats amazing, it triggers the bot in IRC too | |
| 16:15 | [mattermost] <PerplexedTheta> Did we end up deploying this on the customer @mblenk ? If not, something to look at | |
| 16:15 | [mattermost] <PerplexedTheta> Did we end up deploying this on the customer @mblenk ? If not, something to look at / try | |
| 16:15 | [mattermost] <mblenk> I think the best way around it for now until we have a patch is to create a dummy itemtype with the `notforloan` column set to 1 | |
| 16:15 | [mattermost] <mblenk> That should prevent the error appearing, just make sure none of your items are linked to that dummy itemtype | |
| 16:16 | [mattermost] <mblenk> Then when we have a suitable patch you can delete that itemtype | |
| 16:16 | [mattermost] <mblenk> Does that make sense dpk? | |
| 16:18 | dpk | No - I'll need a little more explanation. |
| 16:20 | Our library is a research and archive library so we have lots of items that are not for loan, so this could be relevant. | |
| 16:21 | Need to leave the terminal for about 25 minutes. will check back here when I return. | |
| 16:21 | matterbridge | [mattermost] <mblenk> Do you have any itemtypes in your system that are marked as 'Not for loan'? |
| 16:25 | [mattermost] <mblenk> I'm heading out shortly dpk, I'll be back online tomorrow if you still need help. Have a look at the bug I mentioned above and see if that sounds like the issue you're having | |
| 16:30 | [mattermost] <rtrhimanshu> Hello | |
| 16:31 | [mattermost] <rtrhimanshu> Is anyone live here | |
| 16:33 | [mattermost] <caroline> hello @rtrhimanshu | |
| 16:34 | [mattermost] <rtrhimanshu> Hello | |
| 16:34 | [mattermost] <rtrhimanshu> I was wondering if anyone has tried to bring sort1 and sort2 visible in self registration of patrons | |
| 16:35 | [mattermost] <caroline> I think I remember a bug about that... let me check | |
| 16:35 | [mattermost] <rtrhimanshu> Also is it possible to get automatic data from 264 to 260 | |
| 16:35 | [mattermost] <caroline> [Bug 13901](https://bugs.koha-community.or[…]bug.cgi?id=13901) | |
| 16:35 | huginn` | 04Bug https://bugs.koha-community.or[…]_bug.cgi?id=13901 normal, P5 - low, ---, dudzikmichal, Patch doesn't apply , sort1 and sort2 not on self registration form |
| 16:35 | 04Bug 13901: normal, P5 - low, ---, dudzikmichal, Patch doesn't apply , sort1 and sort2 not on self registration form | |
| 16:36 | matterbridge | [mattermost] <caroline> 264 -> 260 you can use MARC modification templates https://koha-community.org/man[…]ication-templates |
| 16:36 | [mattermost] <caroline> It's not automatic, but it can be used in batch modification, on single records, or when importing MARC files | |
| 16:37 | [mattermost] <rtrhimanshu> that bug is still pending no one has worked on it | |
| 16:37 | [mattermost] <caroline> yep, it happens sometimes... you can comment on it to maybe try to revive it | |
| 16:38 | [mattermost] <rtrhimanshu> I normally use z39.50 to import data so while importing the data gets imported in 264 so i wondered if there is any way to get in 260 | |
| 16:39 | [mattermost] <caroline> hm, I don't think you can use MARC modifiation templates during z39.50 imports, no... someone correct me if I'm wrong. | |
| 16:40 | [mattermost] <caroline> you can save it and then do Edit > Edit with template | |
| 16:42 | [mattermost] <rtrhimanshu> ok | |
| 16:46 | tadzik | is it necessary for the bridge to prefix every IRC message with "[mattermost] "? It's quite obvious that this is where the messages are coming from, and it just adds noise and makes things harder to read |
| 16:48 | dpk | mblenk: Yes, most of our itemtypes (except books) are marked not-for-loan. |
| 16:50 | tadzik: I agree - not sure what the bridge is, but its configuration probably needs tweaking. | |
| 16:52 | Definitely looks like there is config that would help here: https://github.com/42wim/matte[…]EADME.md#settings | |
| 16:53 | Whoever is running the bridge to this channel will need to look into this. | |
| 16:53 | matterbridge | [mattermost] <PerplexedTheta> I can remove the protocol line |
| 16:54 | perplexedtheta joined #koha | |
| 16:55 | matterbridge joined #koha | |
| 16:55 | matterbridge | <PerplexedTheta> Is that better? |
| 16:57 | mattermost joined #koha | |
| 16:57 | ashimema | Matrixx text |
| 16:57 | perplexedtheta | Bad timing! |
| 16:57 | mattermost joined #koha | |
| 16:57 | perplexedtheta | Now try |
| 16:57 | mattermost | [PerplexedTheta] I think that is probably the most concise |
| 16:58 | dpk | Shrink the mattermost nick to mm? |
| 16:59 | We can always see what mm is by doing a right click on mm in the user list. | |
| 17:00 | mattermost | [PerplexedTheta] Seems fair |
| 17:01 | mm joined #koha | |
| 17:01 | PerplexedTheta joined #koha | |
| 17:01 | mm | [PerplexedTheta] Done |
| 17:01 | * mm | [ashimema] considers removing his element client now. |
| 17:01 | dpk | What service are you using matterbridge to gateway? |
| 17:02 | Guest1180 | [PerplexedTheta] The mattermost server, which is hosted in DigitalOcean |
| 17:02 | [ashimema] chat.koha-community.org | |
| 17:03 | [PerplexedTheta] I think we will need to put the nick back to mattermost | |
| 17:03 | [PerplexedTheta] mm is being rejected | |
| 17:03 | dpk | I preferred mm, but as long as its 9 characters or less, you don't mess up the formatting in the IRC client. |
| 17:04 | mattermost is too long. | |
| 17:04 | Guest1180 | [PerplexedTheta] It was working okay in WeeChat |
| 17:04 | [PerplexedTheta] What client do you use? | |
| 17:04 | dpk | XChatAzure on the Mac |
| 17:05 | Guest1180 | [PerplexedTheta] I see - I can try mmost |
| 17:05 | reiveune | bye |
| 17:05 | mmost joined #koha | |
| 17:05 | reiveune left #koha | |
| 17:06 | mmost | [PerplexedTheta] Has that worked? |
| 17:08 | ashimema | mm worked for me too |
| 17:08 | mmost | [PerplexedTheta] It was rejected - probably reserved? Name kept getting changed to Guest |
| 17:09 | [PerplexedTheta] `mmost` should be concise enough hopefully? | |
| 17:09 | dpk | mmost is fine. |
| 17:09 | PerplexedTheta joined #koha | |
| 17:10 | dpk | mmost: mblenk - are you still around? |
| 17:10 | mmost | [ashimema] he'll have finished for the day now dpk |
| 17:11 | [wizzyrea] did someone say that amazon cover images were being a problem today? | |
| 17:11 | dpk | Bummer. |
| 17:11 | wizzyrea: I was commenting earlier that Amazon bots were DDOSing our catalog. | |
| 17:12 | mmost | [wizzyrea] hm ok |
| 17:12 | dpk | The robots.txt file lacked the URL they were usings (they omitted the /cgi-bin/ prefix) |
| 17:13 | mmost | [wizzyrea] we've been getting a lot of alibaba bots recently, just scraping |
| 17:20 | dpk | The searches they are doing are just overwhelming our server's capacity. |
| 17:21 | They are not supposed to be able to do that. | |
| 17:21 | but the robots.txt file was lacking the URLs they were using (E.g. /opac/opac-search.pl vs /cgi-bin/opac/opac-search.pl) | |
| 17:22 | I now have both paths in the file. | |
| 17:22 | But I have to wait for them to re-read the file before it takes effect. | |
| 17:38 | mmost | [rtrhimanshu] Is there a way to import Descriptions of books from amazon |
| 17:41 | [wizzyrea] lots of scrapers don't honour robots.txt. If it's unidentifiable as "helpful" we often block aws IP's | |
| 17:50 | dpk | Amazon explicitly says they will respect robots.txt. I will take that on faith for now. They only read that file |
| 17:50 | once a day and it was clearly missing the URLs they were using. | |
| 17:52 | I thing this may be a bug in the default robots.txt file and related docs for Koha. I suspect the actual URLs in use may have altered over the years, dropping the /cgi-bin prefix, or perhaps it just works with both (have not tested). | |
| 18:09 | mmost | [PerplexedTheta] @wizzyrea we've been getting alibaba stuff to |
| 18:09 | [PerplexedTheta] @wizzyrea we've been getting alibaba stuff too | |
| 18:09 | [PerplexedTheta] I don't think its bots specifically commissioned by alibaba | |
| 18:09 | [PerplexedTheta] I think its people using cloud.alibaba.com for botting | |
| 18:09 | [PerplexedTheta] Probably nation state stuff. Which is worse | |
| 18:10 | [PerplexedTheta] I think its people using alibabacloud.com for botting | |
| 18:13 | [wizzyrea] it's Quite Strange | |
| 18:13 | [wizzyrea] we block it - are almost to the point of blocking their subnets | |
| 18:14 | [wizzyrea] i'm both relieved and unnerved that it's not just us | |
| 18:19 | cait joined #koha | |
| 18:20 | mmost | [ashimema] I suggested the subnet approach.. we haven't quite got to that point yet.. but I think we're close too |
| 19:04 | [PerplexedTheta] We've blocked a few of their subnets | |
| 19:04 | [PerplexedTheta] But haven't gone for entire ASNs yet | |
| 19:04 | wajasu | i was getting slammed every by bots every few months. i updated my robots.txt and it took 4 days before they stopped coming. my CPU dropped to make the opac usable again. that was last November |
| 19:08 | dpk | This is the robots.txt file I currently have: https://dpaste.com/F49PD8WQ3 |
| 19:15 | wajasu | whoa. I'm getting bogged down now as well. I'm investigating. |
| 19:16 | whoa. I'm getting bogged down now as well. I'm investigating. | |
| 19:39 | dpk | Hey, GOOD NEWS. Amazonbot has disappeared from my sight. Looks like they reread the robots.txt file or my email to them was acted on. |
| 19:41 | bag | nice |
| 19:42 | mmost | [caroline] good! |
| 19:48 | wajasu | i reconciled you robots to mine. mine was about the same, but i didn't have the Semrush line and Cache-delay. Mine had the opac-reserve.pl though. i'm about to reload my webserver. |
| 19:56 | dpk | We don't have any reserve traffic - not really that kind of library. Probably reasonable to add it to the list. |
| 19:57 | Now I just want to get to the bottom of this error: | |
| 19:57 | [2024/04/15 20:37:47] [WARN] No method count found for Koha::Items DBIx::Class::SQLMaker::ClassicExtensions::puke(): Fatal: NULL-within-IN not implemented: The upcoming SQL::Abstract::Classic 2.0 will emit the logically correct SQL instead of raising this exception. at /usr/share/koha/lib/Koha/Objects.pm line 583 | |
| 19:57 | at /usr/share/koha/lib/C4/Search.pm line 2049. | |
| 19:58 | Found in plack-opac-error.log | |
| 20:03 | wajasu | dpk: maybe we have the situation for Koha::Items using IN ... NULL and this warning is rewriting accordingly. |
| 20:03 | https://stackoverflow.com/ques[…]ide-not-in-clause | |
| 20:04 | i'm not in the position to look at any code right now. | |
| 20:07 | dpk | This bug was called out earlier as possibly related: https://bugs.koha-community.or[…]_bug.cgi?id=29087 |
| 20:07 | huginn` | 04Bug 29087: major, P5 - low, ---, koha-bugs, NEW , Holds to pull list can crash with a SQL::Abstract puke |
| 22:09 | mmost | [PerplexedTheta] perplexedtheta updated the channel display name from: IRC Bridge to: IRC bridge |
| 22:09 | [PerplexedTheta] perplexedtheta updated the channel purpose to: A bridge between the IRC and Mattermost servers | |
| 22:10 | wajasu | i have this one IP that I suspect is from a crawler, as its mentioned in the 1st colum of plack.log. |
| 22:10 | i put a ufw deny rule for that IP, but i still have GET requests coming in the plack.log | |
| 22:11 | are requests somehow queued to rabbitmq or some other plack mechanism? is that why i still see request soming in. | |
| 22:12 | its keeping my VPS server CPU at 40% so my opac take 60secs to return a page. | |
| 22:13 | mmost | [PerplexedTheta] Do you use cloudflare or something similar? |
| 22:13 | [PerplexedTheta] I got fooled by X-Forwarded-For a while back | |
| 22:15 | wajasu | oh. maybe i need to get the ufw rule before the https access :) |
| 22:33 | well, since i was using ufw, i added: ufw insert 1 deny from <Crawler_IP> to any | |
| 22:34 | did a ufw reload, and tail -f /var/log/koha/<kohainstance>/plack.log and the crawler isn't getting in for now. | |
| 22:49 | pastebot | "wajasu" at 127.0.0.1 pasted "example of IP block with ufw firewall for a crawler. note you must have the DENY before the ALLOW rule" (28 lines) at https://paste.koha-community.org/42128 |
← Previous day | Today | Next day → | Search | Index
