Closed Bug 40191 Opened 24 years ago Closed 4 years ago

Bad 'connection refused' message

Categories

(SeaMonkey :: MailNews: Message Display, defect)

defect
Not set
normal

Tracking

(Not tracked)

RESOLVED INCOMPLETE

People

(Reporter: sfraser_bugs, Unassigned)

References

Details

Attachments

(3 files)

When trying to connect to a mail server that is down, I get an alert that says:

"Connection refused to the server."

This is bad English. Better alternatives:

"Could not connect to the mail server; the connection was refused."
"Your connection was refused by the mail server."

or something.
reassigning to jefft and cc'ing jglick.  FWIW I like Simon's second option.
Assignee: putterman → jefft
It is possible to say WHY the connection was refused? 

"Could not connect to the server <name> because it is currently not available."

Simone?
I agree we should make this error message more friendly (and use better 
English). I think it would be great if we used Jennifer's suggestion (stating 
the reason) and put the name of the server in there as well. Is that possible? 
I am not sure that we can always get a reason from the server. Sometimes the 
network can be saturated or just died. But I like jglick's error message. It 
sounds good to me.
Status: NEW → ASSIGNED
I realize this is late in the game but you guys should look at how the squid
proxy does things. If I get an error squid generates a nice HTML page. Attached
is one for a refused connection. Squid just uses a bunch of template pages so
something similar should be possible for Mozilla
Target Milestone: --- → M17
Moving to M18 and nominating for beta3.
Keywords: nsbeta3
Target Milestone: M17 → M18
QA Contact: lchiang → laurel
- per mail triage
Whiteboard: [nsbeta3-]
Target Milestone: M18 → Future
Adding myself to cc: list.
There should be three parts to any error message: (1) what happened, (2) why it 
happened, (3) what the user can do about it. And you should always be as
context-sensitive as possible.

So for this error, you would have:
------
{mailAppName} could not connect to the server {serverName}, because
{reason, defaulting to "an error occurred"}.

Try again later, or contact the server administrator if the problem
persists.
------
OS: Mac System 8.5 → All
Hardware: Macintosh → All
Matthew's text looks good to me.
reassigning jefft's bugs to naving
Assignee: jefft → naving
Status: ASSIGNED → NEW
Keywords: mail3
marking nsbeta1+
Keywords: nsbeta1
Whiteboard: [nsbeta3-] → [nsbeta3-][nsbeta1+]
moving to mozilla0.8 milestone.
Target Milestone: Future → mozilla0.8
*** Bug 64056 has been marked as a duplicate of this bug. ***
moving to mozilla1.0
Keywords: nsbeta3
Whiteboard: [nsbeta3-][nsbeta1+] → [nsbeta1+][1/15]
Target Milestone: mozilla0.8 → mozilla1.0
marking nsbeta1-
Keywords: nsbeta1nsbeta1-
Whiteboard: [nsbeta1+][1/15] → [nsbeta1+ 1/15]
Target Milestone: mozilla1.0 → Future
I realize this patch won't satisfy the full context-sensitive error message 
hopes, but it is an improvement nonetheless, over the current implementation.
r=hwaara on stephend's change. The reason no one has touched this bug since
mpt's comment is that it is quite hard to implement, at least for those looking
for these kind of bugs to fix.

So, sure, mpt's suggestion is good. But we can still check this patch in the
meantime and then see when that request is fixed. 

Seth, wanna sr=?
(stephend, )this bug already has my r=. replacing review keyword with approval.
Keywords: reviewapproval
sr=alecf
I don't know whether to mark this fixed or not, but to err on the safe side,
I'll checkin my patch, leaving it open for now.  Fix checked in.
Keywords: approvalverifyme
Reassign to stephend, who already fixed this bug. (Maybe you should mark this 
FIXED?)
Assignee: naving → stephend
Hwaara, I was leaving this open to cover MPT's suggestion that the error
messages be as context-sensitive as possible.  Anyhow, if left open, it
shouldn't be re-assigned to me, rather to someone who can implement that. 
Re-assigning to default component owner.
Assignee: stephend → sspitzer
QA Contact: laurel → esther
FYI, This line that you changed in imapMsgs.properties can also be found in 
messenger.properties toward the very end of the file.

Line 170-  103=Connection refused to the server.

This should also be changed if you want to keep the messages consistent.
Agree its an improvement. :-)
Jessica, thanks for picking this up,
r=jglick@netscape.com/stephend@netscape.com.   Just need super-review now. 
Seth, can you provide that, since you're already on this bug?  Thanks.
That last patch was checked in, still I'm not sure this is totally FIXED
(context-error messages would be better).  Leaving open, for now.
See also bug 75977 "connection refused *by* the server".
*** Bug 105131 has been marked as a duplicate of this bug. ***
Citation:

Sending of message failed.

The message could not be sent because connecting to the SMTP server failed. The
server may be unavailable or is refusing SMTP connections. Please verify, that
your SMTP server setting is correct and try again, or else contact your network
administrator.

Resolved fixed?
Prior to receiving the connection refused dialog box, I get the "trying to
connect" dialog box.  I am on a local area network so I am connected, and often
when this dialog box pop's up I can continue to browse the page I was on.  This
does not only happen on connection refusals, I also get his when trying to
download, for example I was on rpmfind.net and got the same thing.
See also bug 152781. "Refused" is inappropriate unless the connection was in
fact refused, as it will cause users to contact their ISP's unnecessarily to
find out why the connection was "refused".
Product: Browser → Seamonkey
Assignee: sspitzer → mail
is this bug and patch obsolete?

reason I say this is the message I get these days is "failed to connect to
server XXXXX"

other bugs that may be obsolete:
Bug 152781 
Bug 75977
Assignee: mail → nobody
Priority: P3 → --
QA Contact: esther → message-display
Whiteboard: [nsbeta1+ 1/15]
Target Milestone: Future → ---
Keywords: verifyme

I was not able to find complained message string in seamonkey-2.53.1.source.tar.xz
fixed? Or wrong search method?

I am going to close this one within 10 days if we do not get indication that the requested improvement still is required.

Untouched for years, no more complains, unclear whether still exists ...

Status: NEW → RESOLVED
Closed: 4 years ago
Resolution: --- → INCOMPLETE
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: