@RbDev wrote:
Hello all.
In a thread, I want to reply for a specific comment, remember you can’t have subthreads, how can this be achieved?
channel - thread
I like cats
I like red cats
I like big cats < ---- I want to reply to this specific comment
I like black cats
Posts: 1
Participants: 1
@GoodBadboY wrote:
Hello,
While having some problem to run the self-hosted push notifications service. I wanted to know what the Mattermost’s JSON payload for an iOS push notification looked like, but unfortunately, I found nothing on the Internet.
So I was just wondering what the JSON payload send by the Mattermost push notifications service looked like for a new post and an edited one.
Also, I want to know if the “content-available” attribute is already present, and if not, if we could add it.
Thanks.
Posts: 1
Participants: 1
@nkumada wrote:
Dear supporters,
I have faced following trouble after carried out the upgrade, I would appreciated it if someone can tell me workaround to fix or avoid this.
Summary
I have just upgraded Mattermost from 5.08 to 5.13 and have been confirming it,
I have found unable to create “New Channel” after upgraded both of web interface and CLI.Mattermost Enterprise Edition
Modern communication from behind your firewall.
Mattermost Version: 5.13.1
Database Schema Version: 5.13.0
Database: mysqlSteps to reproduce
- Log it in as initial user to Mattermost and then tried creating new channel using plus bottom by “PUBLIC CHANNELS”.
- Select “Public” and enter the name of channel to be created and submit it using “Create new channel” bottom, however, screen has not moved to next with displaying “Unable to find the existing channel”.
- and I have found no new channel has been created.
Expected behavior
Created new intended channel, both of web interface and CLI.
Thanks!
Nobu
Posts: 1
Participants: 1
@compuls1v3 wrote:
If you can’t find an answer, prepare a report with the following information:
Which version of Mattermost are you using? Mattermost Version: 5.12.3 Database Schema Version: 5.12.0 Database: mysqlA link to documentation used and the specific step at which you’re failing.
https<colon forward slash forward slash>docs.mattermost.com/install/config-ssl-http2-nginx.html?highlight=nginx%20ssl Step 7. Running "./letsencrypt-auto certonly --standalone" results in an error. Relevant output of error logs in mattermost.log from directory where Mattermost is installed Surround code and console output with backticks (`) for easy reading. Example: Text beginning and ending with backticksPost this information to the Troubleshooting Forum 85 by clicking “New Topic” to ask your question.
Posts: 1
Participants: 1
@GoodBadboY wrote:
Hello,
I am creating a mobile iOS application using Mattermost server and I have one question about the push notification service.After asking the user to allow push notifications, I received a Device Token. I guess that I have to give this token to the Mattermost server or to the Mattermost push notification service but how do I do this? What I am supposed to do with the iPhone device token so that the Mattermost token send push notification to the right iPhone?
Thanks
Posts: 1
Participants: 1
@VoyagesDivins wrote:
Summary
Mattermost do not go further than “Pinging SQL master database” on startup.
Steps to reproduce
Mattermost team 5.12.3
Ubuntu 18.04
Postgresql 10config.json :
“SqlSettings”: {
“DriverName”: “postgres”,
“DataSource”: “postgres://mmuser:<my_password>@localhost:5432/mattermost?sslmode=disable&connect_timeout=10”,I’ve tested db connection with :
psql --dbname=mattermost --username=mmuser --password
I’ve got access to db without any problemThen I tried to start the server with :
sudo -u mattermost ./bin/mattermost
The last message is “Pinging SQL master database”Expected behavior
The server should start or I should get an error.
Observed behavior
My server’s disk got full, so some sevices stopped working. I’ve clean my disk and restart all services but mattermost does not start anymore.
Posts: 1
Participants: 1
@ahmaddanial wrote:
Problem Definition
- The current behavior of the Unread Message Indicator shows the unread messages in an alphabetical order by default regardless of the timing when the messages are received
- This can potentially cause an issue when a user is communicating with multiple users / channels at a time, not knowing which 1-1 / channel has the most recent unread message and which one to respond to first
Suggested Solution
Propose a new unread message indicator feature that allow users to configure it to be sorted alphabetically or based on recency
Posts: 1
Participants: 1
@RbDev wrote:
Hi all,
HTTPS port 443 stopped working, probably lets encrypt expired, now all mobile clients get stuck on startup. no one can access MM.
No error logs. I dont know if MM is trying to renew and is failing or what.
Iit possible to renew the lets encrypt manually? How do you create the cache folder?
Please help.
Cheers
"ServiceSettings": { "SiteURL": "https://mattermost.mydomain.com", "WebsocketURL": "", "LicenseFileLocation": "", "ListenAddress": ":443", "ConnectionSecurity": "TLS", "TLSCertFile": "", "TLSKeyFile": "", "TLSMinVer": "1.2", "TLSStrictTransport": false, "TLSStrictTransportMaxAge": 63072000, "TLSOverwriteCiphers": [], "UseLetsEncrypt": true, "LetsEncryptCertificateCacheFile": "./config/letsencrypt.cache", "Forward80To443": true,
Posts: 3
Participants: 2
@netgineer wrote:
Hey.
I want to use the Custom Attributes Plugin plugin on my server so that users can fill in custom fields about themselves, such as phone, position, etc.
Do I understand correctly that right now users can’t enter such details on their own, and it must be done by the admin for each user in the
config.jsonfile?In the examples on the GitHub plugin page I see the
userIDparameter - is this the sameuserIDthat I can get in theidline after executingmattermost user search <nickname>?
Posts: 3
Participants: 2
@patrickoman wrote:
I configured “#release” as a word that triggers mentions but I don’t receive notifications if a message contains this or other hashtags. trigger words like “release” and “&release” are working.
are hashtags not supported for mentions or is there a way to use them this way?cheers
Posts: 2
Participants: 2
@marion wrote:
Summary
Upgraded to 5.12.4 and Emoji Picker screen freezes Mac desktop client running Mojave
Steps to reproduce
5.12.4 on RHEL7 vm. Users are using 4.2.1 Desktop Mac client on Mojave.
Expected behavior
User clicks on Emoji picker screen. Mouse should move freely between emojis and then you can click on an emoji and it shows up in your chat bar.
Observed behavior
User clicks on Emoji picker screen. Move mouse over emoji’s and it is very slow with lag. Will then randomly freeze desktop client. Have to go to toolbar and Clear Cache & Reload to get it to come back.
Posts: 2
Participants: 2
@Spartacus2018 wrote:
Hello there,
so im using Mattermost, i create channel, but how adding role ?for now all member can create Channe ,private group channel, and i don’t want this.
I want create for evry channel a teamLeader who can manage user and group discution by channel, how do that please ?
Posts: 2
Participants: 2
@Brownr wrote:
HC2MM Import tool getting hung on C:\xxx\xxx\hipchat_export\users\xxxx\history.json
While running HC2MM from a Windows 10 Desktop PC, it gets hung on a .json file and errors out about 10 minutes after (Error Reading Properties: Stream Closed). Before erroring out it has passed through thousands of json files already but gets hung on one, I have tried several times to remove the troublesome file but it gets hung a few more down the line.
Does anyone know where to find more logs that may point me to why its crashing?
Also it seems like there is quite a bit more info on the hipmost tool, is this the better way to import?
After the tool crashes, would there residual files I should cleanup?
Any help/pointers would be greatly appreciated.
Atlassian HipChat 2.4.1 (2018.07.19.134050)
Mattermost Version: 5.12.3
Database Schema Version: 5.12.0
Database: mysql
Posts: 2
Participants: 2
@JohnWick wrote:
In the default search backend (SQL, the only one available on Team Edition), searching only returns 100 results max.
This seems a bit at odds with the feature comparison table which proudly advertises “Unlimited search history and integrations” for all editions (including Team Edition).
Of course this isn’t an issue for those on Enterprise using the elasticsearch backend, but would you please consider supporting paging or better yet infinite scroll for the poor Team Edition users? Given the wording in the feature comparison table (and how the code looks), I personally feel that this was an unintentional oversight, but it is nonetheless a rather significant one.
There is already an existing feature suggestion, so I suppose this is a cross-post. If you are receptive to this idea, please consider adding some votes or discussing it.
Thank you for your time.
Posts: 1
Participants: 1
@yanuk wrote:
I am running MM server 5.13.0 and Android app version 1.21.1 on a Samsung Galaxy Note 5 running Android 7.0
The same phone was able to connect using 1.21.0 but after updating to 1.21.1 it stopped connecting. This is the only phone in office that doesn’t connect. Older phones on Android 5.0 using MM App 1.14 was able to connect. Is there any issue with the Android OS requirement?
Posts: 1
Participants: 1
@alex_enyway wrote:
'm building a slash command. The flow I imagine is:
- User triggers slash command in client
- My server handles request and returns interactive dropdown
- User selects option from dropdown in client
- My server handles the POST request from the selection and returns an interactive button
- User presses the button
Something seems to go wrong between steps 4 and five. The server handles the selection and returns the desired response:
{ "update": {"message": "Updated!"}, "ephemeral_text": "You updated the post!", "attachments": [{ "text": "Ready?", "actions": [{ "name": "Go!", "integration": { "url": "https://somewhere.com?foo=bar", "context": {"action": "event_submitted"} } }] }] }…but nothing shows up in the client.
Is something wrong with that JSON? I have tried other things like only returning a text message.
I’ve set log level to DEBUG and don’t find anything suspicious there.
I wonder, if I have misunderstood the docs and the client will never even evaluate a response from a POST triggered by an interactive message. In that case I would have to make a new request back from my server to the Mattermost API in order to get to the next step in my workflow. Is that correct?
I have asked the same question on StackOverflow, but no answers yet.
Posts: 1
Participants: 1
@Hardik1705 wrote:
Hi,
Can anyone help me with the code to upload a csv and sent it as a message using API?
Posts: 1
Participants: 1
@mmfan wrote:
Summary
I’ve had Mattermost configured with an Apache reverse proxy for a few weeks, so visiting https://mattermost.domain.tld resolved just fine. Mattermost is listening on port 9065. It seems that after enabling the API option in the Mattermost console somehow tripped up Mattermost. Even after disabling this option for a few days going to mattermost.domain.tld with https results in this message:
Client sent an HTTP request to an HTTPS server.I cannot for the life of me figure out why this keeps happening. I have several other Apache VirtualHosts defined for other services using a reverse proxy without trouble, so I’m not sure anymore why Mattermost just won’t cooperate. I can visit mattermost.domain:9065 specifying the port with https, this does resolve fine. Up until the point it began giving the above client message my Apache config was not changed. I didn’t make any changes in Mattermost except for enabling then disabling an option related to the Mattermost API, but I don’t see how this would be causing this.
Steps to reproduce
Mattermost version 5.12.2
Apache version 2.4.39Expected behavior
This is the original Apache config I have been using just fine with Mattermost by visiting only mattermost.domain.tld (https) without specifying port 9065, but now gives the client https message.
<VirtualHost *:80> ServerAdmin webmaster@domain ServerName mattermost.domain.tld ServerAlias mm.domain.tld Redirect permanent / mattermost.domain.tld <--https </VirtualHost> <VirtualHost *:443> ServerName mattermost.domain.tld ServerAlias mm.domain.tld ServerAdmin webmaster@domain ProxyPreserveHost On # Set web sockets RewriteEngine On RewriteCond %{REQUEST_URI} /api/v[0-9]+/(users/)?websocket [NC,OR] RewriteCond %{HTTP:UPGRADE} ^WebSocket$ [NC,OR] RewriteCond %{HTTP:CONNECTION} ^Upgrade$ [NC] RewriteRule .* ws://127.0.0.1:9065%{REQUEST_URI} [P,QSA,L] Redirect permanent / mm.domain.tld <---https:// <Location /> Require all granted ProxyPass 127.0.0.1:9065/ <--http ProxyPassReverse 127.0.0.1:9065/ <--http ProxyPassReverseCookieDomain 127.0.0.1 mm.domain.tld </Location> SSLEngine on SSLProtocol all -SSLv2 -SSLv3 SSLCertificateFile "/etc/letsencrypt/live/domain.tld/cert.pem" SSLCertificateKeyFile "/etc/letsencrypt/live/domain.tld/privkey.pem" SSLCACertificateFile "/etc/letsencrypt/live/domain.tld/fullchain.pem" </VirtualHost>I have scrounged the Mattermost forums, GitHub and other similar Apache reverse proxy problems, but no matter how I configure Apache loading https://mattermost.domain.tld still results in the dreadful
Client sent an HTTP request to an HTTPS server.Observed behavior
Checking the Mattermost logs doesn’t show any errors or messages that would indicate a problem with Mattermost. The Apache error logs don’t show any problem.
Since I am using an Apache reverse proxy I have also ensured
UseLetsEncryptandForward80To443are false.Note that I have https before the domain below, but I can’t post more than 2 links.
"SiteURL": "mattermost.domain.tld", "WebsocketURL": "", "LicenseFileLocation": "", "ListenAddress": ":9065",The Mattermost directory
/usr/share/webapps/mattermostis also owned by the mattermost user and group. Considering I didn’t change anything in Apache or Mattermost related to SSL/TLS options I am at a complete loss on how to fix this. I’ve tried restarting Mattermost and Apache for good measure and following the logs, but still nothing makes sense to me why I’m getting theClient sent an HTTP request to an HTTPS servermessage.
Posts: 2
Participants: 1
@Taffer wrote:
The current process for upgrading a Mattermost server (found here) is somewhat painful, and feels like it takes longer than it should.
Ideally, I’d like to be able to trigger a server upgrade from the System Console, and have it take care of things for me. The same goes for server plugins… currently the administrator needs to watch the GitHub page for every plugin (or follow @mattermost onTwitter). The same goes for the desktop Mattermost client. Mobile clients get automatically updated whenever a new version is released (perfect!).
What needs to be done to make this happen?
The desktop client is probably fairly straightforward… things like Discord and Slack tell you when you need to update, you click a thing, it updates. Maybe Electron has a canned API for this, or maybe there’s a commonly-used technique?
For plugins, the server code will have to have a way to see what version(s) are available, and (presumably) what server API level they require.
For the server, I have no idea. Split plugins into “comes with the server code” vs. “added later by admins”, to make it safe® to update? Move user data out of the
mattermostdirectory completely? Do other Go-based servers have a good method for handling this?
Posts: 1
Participants: 1
@leosimony wrote:
One of our users complained that,
A direct message from one user was not landed in Mattermost App but received an email notification for the same. Once he logout and login the message appeared in the Mattermost app showing as landed correctly.
Anyone else faced this scenario before?
Mattermost Server version 5.12.2
Posts: 2
Participants: 2