What web server are you utilizing? And have you analyzed the returned http
headers to ensure your issue is the same as I was encountering? (If you
send me the link to your site, I can check them out for you.)

-Chris


On Tue, Jan 10, 2017 at 6:23 AM, Gianluca Pascarella <
[email protected]> wrote:

> Hi, I've the same problem, unfortanately I'm not able enought to
> understand and fix it as you did, can you describe what should I do? Thanks
> a lot in advance, best regards
>
> Il giorno sabato 10 dicembre 2016 06:49:25 UTC+1, Chris Mattingly ha
> scritto:
>>
>> Ahhh... it paid to do a request/response header comparison between your
>> site on mine. This revealed that nginx was setting the 'Cache-Control:
>> max-age' value to some insanely high number (due to the default nginx
>> setting of 'expires max' for most media type files).
>>
>> I removed the ?<seconds since epoch> query parameter from the templates
>> for my graphs, created an exception in nginx so it no longer sets that
>> header for png files from my weather base-url, and bingo - the graphs are
>> refreshing correctly.
>>
>> Case closed! :)
>>
>> -Chris
>>
>>
>> On Sat, Dec 10, 2016 at 12:01 AM, Andrew Milner <[email protected]>
>> wrote:
>>
>>> http://cr.netflash.com.cy:61016/weewxweather/Bootstrap/index.html
>>>
>>>
>>>
>>> On Saturday, 10 December 2016 06:41:59 UTC+2, Chris Mattingly wrote:
>>>
>>>> It doesn't make sense that it'd be two different browsers behaving the
>>>> same. My ISP does no caching, and it wasn't refreshing whether I'm going
>>>> through my ISP or cellular.
>>>>
>>>> What's your site? I could give that a test and see what happens over a
>>>> few days' worth of reloading the page.
>>>>
>>>>
>>>> On Fri, Dec 9, 2016 at 11:37 PM, Andrew Milner <[email protected]>
>>>> wrote:
>>>>
>>>>> My phone (not iPhone) always gives me the latest images and always
>>>>> refreshes when I press browser reload.  Sounds like a quirk of iPhone or
>>>>> maybe your ISP ....
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Saturday, 10 December 2016 06:23:25 UTC+2, Chris Mattingly wrote:
>>>>>
>>>>>> I already have auto-refresh on the page and this is not a problem on
>>>>>> "real" browsers - just on the mobile devices. Again, I have a workaround 
>>>>>> in
>>>>>> place, but was just looking for other (better) solutions.
>>>>>>
>>>>>> I skimmed through multiple hits on this problem over at stackoverflow
>>>>>> and most point to this same "cache-busting" fake query parameter to the 
>>>>>> img
>>>>>> src tag.
>>>>>>
>>>>>> The real solution here should be to use unique filenames, which
>>>>>> understandably won't work for everyone (e.g. not using the rsync delete
>>>>>> option), but would be better in my situation so that the cache can be
>>>>>> properly used for the less-frequently updated images. Sounds like a good
>>>>>> project to avoid the family over the upcoming holiday. ;)
>>>>>>
>>>>>> -Chris
>>>>>>
>>>>>>
>>>>>> On Fri, Dec 9, 2016 at 10:52 PM, Andrew Milner <[email protected]
>>>>>> > wrote:
>>>>>>
>>>>>>> Images should be reloaded when refresh button is pressed in the
>>>>>>> browser if they have been updated whilst browser is open on the page.  
>>>>>>> Or
>>>>>>> are you seeking an auto-refresh capability?
>>>>>>>
>>>>>>>
>>>>>>> On Friday, 9 December 2016 23:27:26 UTC+2, Chris Mattingly wrote:
>>>>>>>
>>>>>>>> I've searched the archives with no luck for this issue.
>>>>>>>>
>>>>>>>> When a given skin doesn't change the filename for the graphs
>>>>>>>> (day*.png, etc.) the images are never updated/reloaded when viewing on
>>>>>>>> mobile browsers (tested Safari & Chrome on an iPhone).
>>>>>>>>
>>>>>>>> My workaround was to append '?$current.dateTime.raw' to the img
>>>>>>>> src. The minor downfall with this is that for the images that are not
>>>>>>>> updated every report cycle (weekly/monthly), since the html file is 
>>>>>>>> still
>>>>>>>> regenerated, the current timestamp will still be appended. Granted, 
>>>>>>>> since
>>>>>>>> these images are so small, it's not much to worry about.
>>>>>>>>
>>>>>>>> Are there any better fixes for this problem? Given it's impossible
>>>>>>>> to know which browsers will obey any of the Cache/Pragma/Expires
>>>>>>>> parameters, maybe there isn't...
>>>>>>>>
>>>>>>>> Thanks!
>>>>>>>>
>>>>>>>> -Chris
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>> You received this message because you are subscribed to the Google
>>>>>>> Groups "weewx-user" group.
>>>>>>> To unsubscribe from this group and stop receiving emails from it,
>>>>>>> send an email to [email protected].
>>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>>>
>>>>>>
>>>>>> --
>>>>> You received this message because you are subscribed to the Google
>>>>> Groups "weewx-user" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>>> an email to [email protected].
>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>
>>>>
>>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "weewx-user" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to [email protected].
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>
>> --
> You received this message because you are subscribed to the Google Groups
> "weewx-user" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to