I followed the instructions at https://developers.google.com/compute/docs/sending-mail to setup postfix to send via sendgrid. I believe I have it all setup correctly, but mail is failing. The bounce is not very helpful other than status: 5.0.0 and Action: failed.
Any ideas?
As a wild guess, you might not have run postmap from step 5, or reloaded postfix after updating your /etc/postfix/main.cf. Can you post the log messages from postfix from /var/log/syslog along with your postfix main.cf?
One other question: are you running on Debian, CentOS or something else? I think that the SendGrid instructions you linked to assume Debian, and file locations may be different in CentOS.
Related
After installing Postfix on my Ubuntu server and getting my Symfony application running, I'm attempting to have the system send emails via Postfix. I tested it out in the terminal to make sure that it sends properly and it does, but I'm confused as to what the Symfony configurations are to have SwiftMailer send email through Postfix.
In the .env file, there is a MAILER_URL attribute. What do I input there? My hostname? One of the destinations? Something else entirely? Help would be appreciated.
I use Symfony and SwiftMailer as post manager. I can send post using console command, but when I send using browser client the post is not sent and there is nothing in log files. What can be the reason of such strange behavior?
CLI - Command Line Interface
CGI - Common Gateway Interface
You probably mean CGI php that deals with browser routine.
I faced the same problem using nginx. Swiftmailer refuse to send any e-mails triggered by CGI PHP, but nevertheless it successfully sends e-mails by swiftmailer:email:send command in console.
I found that the cause of this problem is function stream_socket_client in disabled_functions of my nginx config. Just remove stream_socket_client from disabled functions, reboot your server and try again, it should help.
I am building a custom slash command for slack. When the slack user types a command, ex /uptime, a HTTP POST message is sent to the server URL.
The tutorials I've read all include installing a tool such as ngrok, pagekite, or localtunnel to generate a URL for the local machine.
Since I am working with a server, can I not just open a port and have slack connect directly to that hostname and port? How can I do this?
Doing some research, I came across opening a port with nc, then listening with curl, however I don't understand how to put it all together.
Yes, if you are running your script for handling the POST requests from Slack on a server that has a URL that can be reached on the Internet you do not need a local tunnel like ngrok.
If you starting from scratch I can recommend using a standard Apache + PHP [+ MySql] stack and have a PHP script to interpret and react to the POST request. Of course other script languages (e.g. Python) work just as well.
I am trying out Eldos's SFTP Net Drive to map a drive to a virtual Ubuntu server. I works great when authenticating with a password. I have tried the normal/typical methods for configuring Key-based access. However, I receive error 10058. I have searched for clear instructions without success.
Anyone using this with Key-based access? Please share how you have it configured.
Thank you :)
If you're connecting via command line (e.g. using "open /profile:server" command line arguments) and you're accessing recently reinstalled server, the connection will silently fail because of changed server ssh fingerprint.
Try to reconnect manually (not using command line), accept new fingerprint and the problem will resolve.
This is because the server doesn't accept the key for whatever reason and closes connection. Usually this is an indicator of buggy server which normally should send the error packet in response.
But your question is offtopic here on StackOverflow.
I've run into an issue configuring a Yum repository which I'm serving up with HTTP/Apache.
I want to use http basic authentication, and set the listening port to something other than 80, because this repo is in my DMZ. I was trying to do that by setting up my baseurl as follows (from my repo file):
baseurl=http://user:password#<my-ip>:<my-non-80-port>/yum/RHEL/7/base/$basearch/
The documentation for yum.conf indicates you can do basic http authentication by adding the username/password to the URL as I did above. I get the following error when I try to do a yum search mysql from a client:
[Errno 14] HTTP Error 401: Authorization Required
This indicated to me that the extra ":" specifying the port was confusing yum's parsing of my url. I removed the port, set the server back to port 80, and it works.
So it appears you cannot do basic http authentication AND change the port number that the web server listens on. I noticed that there are username and password directives in newer version of Yum, but my version doesn't appear to have this directives (I am running yum-3.2.22-37.0.1.el5). Is there any workaround for this?
I worked around this by turning off plugins by adding the following line to my yum configuration.
plugins=0
I think the workaround worked because the base yum updater was working fine with the authorization information from baseurl, but some of the plugins (such as the fastmirror plugin) didn't remember the authorization properly. So the behaviour I had without plugins enabled is that initially an update would work, then subsequently it would start to fail with the authorization error in the question. Once I disabled plugins, everything started to work fine.