All tgcf.config.yml content are gone!

[simonlkch]

I create tgcf.config.yml and excute by the docker i forward 15 channel one by one, and this is past mode `

• source: -1001206172900 dest:
  • -1001174762000 offset: 1792 end: 0
• source: -1001157449000 dest:
  • -1001506569000 offset: 1874 end: 0
• source:

go on......

`

all the config is good and the program run smoolty. but today i start the docker run again. all of the config content are gone, the yml file is empty.

so the offset are all wrong, cause program will forward the duplicate message.

Suggestion

Does the program have any log file to check the last offset? or the offset number write in to other file, and don't replace the yml file?

[tissole]

This is strange! I know that on Heroku all the changes to the config file (the offset value) are lost after the dyno restarts. But it should not erase the whole file. Where are you running the script? If you run tgcf past --loud more details will appear in the console.

[simonlkch]

@tissole for this comment tgcf past --loud, it show the message which is on the console right now. that means that is no previous message.

i use the program by docker, after docker close the message will re-write

[tissole]

So now is working correctly (it starts from msg-id 94 because you set 93 as the offset value)? And after you close it the config file gets erased?

[simonlkch]

So now is working correctly (it starts from msg-id 94 because you set 93 as the offset value)? And after you close it the config file gets erased?

i create a new config yml and re-config the yml, yes i set it to 93. After close the docker, and re-run. the yml still here

i can't Reproduce the case.

for more detail:

before the config file become empty. the program have a lots of error, all of the eror said that can't send to the dest channel on console. but the message could send to the channel. After this situation run a few day, then the docker are stop, i re-run the docker, the config file become empty

[tissole]

Strange indeed! I don't know what is happening, let's hope @aahnik returns soon and fixes this issue.