Forum Replies Created

Viewing 15 replies - 1 through 15 (of 18 total)
  • Thread Starter sanderpinkse

    (@sanderpinkse)

    Oops… Please disregard everything I just wrote. I was having a brain meltdown.

    I expected the new comment notifications to work the same way as before, but they are of course only triggered by comments that DON’T need moderation. Quite logical, otherwise you would receive two notifications for some comments.

    Sorry about the false alarm, and creating unnecessary confusion…

    Thread Starter sanderpinkse

    (@sanderpinkse)

    It is odd, especially because the comment notifications worked fine before I upgraded today – and as I said, comment moderation notifications do work.

    I tried different email adresses as well (I’m using your Send to Any Email add-on), just to make sure it wasn’t something to do with email deliverability.

    And no, no smtp of email log plugins involved.

    I’m also using your Custom Fields add-on. Maybe that could be a factor here?

    Thread Starter sanderpinkse

    (@sanderpinkse)

    I did some exploring. Turns out my server is running PHP 5.6, but the WordPress application was still on PHP 5.4. I figured out how to change that to PHP 5.6, and upgraded to Rsizefly 2.0.1 without any problems. Thanks!

    The new options look very promising, btw.

    Cheers!

    Thread Starter sanderpinkse

    (@sanderpinkse)

    Hi Alex,

    Thanks for the quick reply!

    I just SSH-ed into my server:

    PHP 5.6.32 (cli) (built: Oct 28 2017 06:26:52)

    Strange…

    Wow! That was interesting timing. Jetpack 5.4 (released minutes ago) solved my problem. Cheers!

    I think it is a caching problem, in combination with the iframe. When I open a page in a private or incognito window, the comments form is fine. In a regular window, it displays the same problems as described here (and the button to submit comments is pushed out of sight when the comment is too long).
    I’m using the Sucuri firewall, which probably is a factor as well.

    Thread Starter sanderpinkse

    (@sanderpinkse)

    Ah! That makes sense. Thanks a lot for the quick response!

    Thread Starter sanderpinkse

    (@sanderpinkse)

    This is what I’m seeing with JS optimization on:

    Uncaught ReferenceError: Parsley is not defined
    at autoptimize_5d5952f….js:321
    (anonymous) @ autoptimize_5d5952f….js:321

    And this when I turn it off:

    Uncaught ReferenceError: Parsley is not defined
    at nl.js?ver=1.5.0.4:4
    (anonymous) @ nl.js?ver=1.5.0.4:4

    In the first case the Caldera modals are not working, in the second case they are…

    Cheers!

    Thread Starter sanderpinkse

    (@sanderpinkse)

    Modals are working for me again as well – but only if I disable the Autoptimize plugin, that minifies JS.

    Thread Starter sanderpinkse

    (@sanderpinkse)

    Sorry, because of the weekend it took a bit longer to get back to you on this.

    This is from the Javascript console on Chrome. The first two are present if I load a page with a Caldera shortcode, the last one appears after I click the button. Probably more than you need…

    GET https://[xxx]/wp-content/plugins/caldera-forms/assets/build/js/vue/vue-filter.min.js 404 (Not Found)
    send @ jquery.js?ver=1.12.4:4
    ajax @ jquery.js?ver=1.12.4:4
    n._evalUrl @ jquery.js?ver=1.12.4:4
    Ha @ jquery.js?ver=1.12.4:3
    append @ jquery.js?ver=1.12.4:3
    (anonymous) @ formobject.min.js?ver=1.5.0.3:1
    (anonymous) @ formobject.min.js?ver=1.5.0.3:1

    Uncaught SyntaxError: Unexpected token <
    at eval (<anonymous>)
    at jquery.js?ver=1.12.4:2
    at Function.globalEval (jquery.js?ver=1.12.4:2)
    at text script (jquery.js?ver=1.12.4:4)
    at Xb (jquery.js?ver=1.12.4:4)
    at y (jquery.js?ver=1.12.4:4)
    at c (jquery.js?ver=1.12.4:4)
    at Object.send (jquery.js?ver=1.12.4:4)
    at Function.ajax (jquery.js?ver=1.12.4:4)
    at Function.n._evalUrl (jquery.js?ver=1.12.4:4)
    (anonymous) @ jquery.js?ver=1.12.4:2
    globalEval @ jquery.js?ver=1.12.4:2
    text script @ jquery.js?ver=1.12.4:4
    Xb @ jquery.js?ver=1.12.4:4
    y @ jquery.js?ver=1.12.4:4
    c @ jquery.js?ver=1.12.4:4
    send @ jquery.js?ver=1.12.4:4
    ajax @ jquery.js?ver=1.12.4:4
    n._evalUrl @ jquery.js?ver=1.12.4:4
    Ha @ jquery.js?ver=1.12.4:3
    append @ jquery.js?ver=1.12.4:3
    (anonymous) @ formobject.min.js?ver=1.5.0.3:1
    (anonymous) @ formobject.min.js?ver=1.5.0.3:1

    Uncaught TypeError: Cannot read property ‘open’ of undefined
    at HTMLButtonElement.eval (eval at <anonymous> (jquery.js?ver=1.12.4:2), <anonymous>:1:7215)
    at HTMLDocument.dispatch (jquery.js?ver=1.12.4:3)
    at HTMLDocument.r.handle (jquery.js?ver=1.12.4:3)
    (anonymous) @ VM160:1
    dispatch @ jquery.js?ver=1.12.4:3
    r.handle @ jquery.js?ver=1.12.4:3

    Thread Starter sanderpinkse

    (@sanderpinkse)

    Hi Jack,

    I did switch to plain text, because another three notifications were markled as psam:

    X-Spam-Score: 3.031
    X-Spam-Level: ***
    X-Spam-Status: Yes, score=3.031 tagged_above=-999 required=3
    	tests=[HTML_MESSAGE=0.001, HTML_MIME_NO_HTML_TAG=0.635,
    	MIME_HTML_ONLY=1.105, RCVD_IN_DNSWL_LOW=-0.7,
    	TO_NO_BRKTS_HTML_ONLY=1.999, T_RP_MATCHES_RCVD=-0.01,
    	URIBL_BLOCKED=0.001]

    Plain text does produce rather ugly mails, though:

    <p><strong>This is a test email. All shortcodes below will show in place but not be replaced with content.</strong></p>
    <p>[comment_author] ([comment_author_email]) schreef:</p>
    <p><em>[comment_content]</em></p>
    <p><strong>Comment [comment_approved]</strong></p>
    <p>[permalink]</p>

    Shouldn’t these tags get stripped out?

    Cheers,

    Sander

    Thread Starter sanderpinkse

    (@sanderpinkse)

    A quick follow up: I just received another notification:

    X-Spam-Status: No, score=0.392 tagged_above=-999 required=3
    	tests=[HTML_MESSAGE=0.001, HTML_MIME_NO_HTML_TAG=0.377,
    	MIME_HTML_ONLY=0.723, RCVD_IN_DNSWL_LOW=-0.7, T_RP_MATCHES_RCVD=-0.01,
    	URIBL_BLOCKED=0.001]

    No mention of TO_NO_BRKTS_HTML_ONLY at all…

    SpamAsassin has me at a loss. Please don’t spend any more time on this…

    Cheers,

    Sander

    Thread Starter sanderpinkse

    (@sanderpinkse)

    Hi Jack,

    That’s right. More specifically:

    TO_NO_BRKTS_HTML_ONLY To: misformatted and HTML only

    I think if you would supply a text/plain alternative besides text/html, the overall score would be much lower.

    Cheers,

    Sander

    Thread Starter sanderpinkse

    (@sanderpinkse)

    Hi Jack,

    I’m not quite sure that is true. The two mails I quoted form above were sent using the same From: address. One of them scored 0.49, the other 2.391.

    However, that is not my point. Apparently, any html mail that is sent using wp_mail() invokes the TO_NO_BRKTS_HTML_ONLY rule in SpamAssassin. That rule has a weight of 2, so I assume you would want to avoid it.

    I can easily work around the problem (changing the From: address, not sending as html), but you probably want BNFW to work properly using the default setup.

    Cheers,

    Sander

    Thread Starter sanderpinkse

    (@sanderpinkse)

    Hi Jack,

    That is correct. I changed the From address to an email address on the same domain that I also use for different purposes. I think SpamAssassin recognises that as a “known sender”, or something like that.

    Here are the relevant headers (I used the Send Me a Test Email option in BNFW):

    X-Spam-Status: No, score=0.49 tagged_above=-999 required=3
    	tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, HTML_MIME_NO_HTML_TAG=0.377,
    	MIME_HTML_ONLY=0.723, RCVD_IN_DNSWL_LOW=-0.7,
    	TO_NO_BRKTS_HTML_ONLY=1.999, T_RP_MATCHES_RCVD=-0.01]

    Hold on… I just repeated that test, and now the scores are high again:

    X-Spam-Status: No, score=2.391 tagged_above=-999 required=3
    	tests=[HTML_MESSAGE=0.001, HTML_MIME_NO_HTML_TAG=0.377,
    	MIME_HTML_ONLY=0.723, RCVD_IN_DNSWL_LOW=-0.7, TO_NO_BRKTS_HTML_ONLY=2,
    	T_RP_MATCHES_RCVD=-0.01]

    Just goes to show how little I understand of the workings of SpamAssassin.
    I have no idea why the Bayesian filter kicks in on the first test, but not on the second…

    Cheers,

    Sander

Viewing 15 replies - 1 through 15 (of 18 total)