Split body of mails not respecting RFC2822 #49
Conversation
|
If we split some lines, I think we should split all lines. For simplicity's sake, I'd split them strictly at byte 990 or so... |
|
So you want me to remove the discovery of the space and always split long lines? |
|
done |
| if (!had_headers) { | ||
| if (linelen > 1000) |
There was a problem hiding this comment.
shouldn't we also break the header lines? that would be complicated though.
There was a problem hiding this comment.
Yes I followed the discussion on the ticket and decided to make a first step splitting the body
| @@ -425,36 +442,44 @@ readmail(struct queue *queue, int nodot, int recp_from_header) | |||
| } | |||
| } | |||
|
|
|||
| if (strcmp(line, "\n") == 0 && !had_headers) { | |||
| if (line[0] == '\n' && !had_headers) { | |||
| } | ||
| if (!nodot && linelen == 2 && line[0] == '.') | ||
| break; | ||
| if (!nocopy) { | ||
| if (fwrite(line, strlen(line), 1, queue->mailf) != 1) | ||
| return (-1); | ||
| if (newline[0] != '\0') { |
There was a problem hiding this comment.
I don't like this conditional - it is very implicit.
For mails where the body does not respect RFC2822, split at the arbitrary length of 990 This should address corecode#18
|
I think I addressed all the comments |
|
Ping? |
|
We ran into unexpected issues that this would fix when trying to switch to dma. For instance, cron jobs would fail if there was a long line on stdout/stderr as it gets piped to sendmail. |
|
Just wanted to comment that while this PR seems to still apply cleanly, when we deployed it we instantly had complaints about some automated emails getting rejected with a corrupted queue error, so failing this check: https://github.com/corecode/dma/blob/master/net.c#L582 I sadly never got a copy of one in order to replicate the issue on-demand, so we just reverted the patch for now. If I find a way to replicate the issue, I'll dig some more into this. |
|
Hi @jailbird777 Regarding your comment on on 20 Feb 2020, can you try to replicate your issue with instructions on FreeBSD PR266629 ? And, if running FreeBSD, you can also try corresponding patches. |
For mails where the body does not respect RFC2822, try to split by words
finding the last space before 1000's character
If no spaces are found then consider the mail to be malformed anyway
This should address #18