Weird time delay in data.iana.org 2024b update
It appears to me that where on network you access data.iana.org <https://data.iana.org/time-zones/releases/> from might make a difference in what content you get at a given time. Are there mirrored servers that perhaps get delayed updates? I have an automated process that looks for new tzdata releases by periodically checking https://data.iana.org/time-zones/releases/. I had received the mailing from this list that 2024b had been released, but had no sign from my automated process that the update had occurred. My polling interval is 30 minutes, so detection should be fairly fast. If used my own web browser at home to access https://data.iana.org/time-zones/releases/ I could see 2024b. Oddly enough, however, when I SSHed the remote server where my process runs, and used curl at the command line to check https://data.iana.org/time-zones/releases/, 2024b wasn’t there yet. I wondered if my server had simply cached old data, but after a reboot of the server the same stale data came through without 2024b being shown as available. It was at least a full day, maybe more, before 2024b was visible from that remote server. Is this perhaps deliberate, to prevent too many people trying to access updated timezone data all at once?
On 2024-09-06 16:42, Kerry Shetline via tz wrote:
It appears to me that where on network you access data.iana.org <https://data.iana.org/time-zones/releases/> from might make a difference in what content you get at a given time. Are there mirrored servers that perhaps get delayed updates?
I have an automated process that looks for new tzdata releases by periodically checking https://data.iana.org/time-zones/releases/ <https://data.iana.org/time-zones/releases/>. I had received the mailing from this list that 2024b had been released, but had no sign from my automated process that the update had occurred. My polling interval is 30 minutes, so detection should be fairly fast.
If used my own web browser at home to access https://data.iana.org/time-zones/releases/ <https://data.iana.org/time-zones/releases/> I could see 2024b. Oddly enough, however, when I SSHed the remote server where my process runs, and used curl at the command line to check https://data.iana.org/time-zones/releases/ <https://data.iana.org/time-zones/releases/>, 2024b wasn’t there yet. I wondered if my server had simply cached old data, but after a reboot of the server the same stale data came through without 2024b being shown as available.
It was at least a full day, maybe more, before 2024b was visible from that remote server.
Is this perhaps deliberate, to prevent too many people trying to access updated timezone data all at once?
More likely that server or org network access is front-ended by a proxy server that has to scan everything before it caches it and/or allows anyone to access it. -- Take care. Thanks, Brian Inglis Calgary, Alberta, Canada La perfection est atteinte Perfection is achieved non pas lorsqu'il n'y a plus rien à ajouter not when there is no more to add mais lorsqu'il n'y a plus rien à retirer but when there is no more to cut -- Antoine de Saint-Exupéry
That is correct. Unfortunately the CDN cache was not purged at the time of deployment, as is usual practice. I apologize for the confusion. Kind regards, James Mitchell IANA From: Brian Inglis via tz <tz@iana.org> Organization: Systematic Software Reply-To: "tz@iana.org" <tz@iana.org>, "brian.inglis@systematicsw.ab.ca" <brian.inglis@systematicsw.ab.ca> Date: Friday, September 6, 2024 at 4:47 PM To: "tz@iana.org" <tz@iana.org> Subject: [tz] Re: Weird time delay in data.iana.org 2024b update On 2024-09-06 16:42, Kerry Shetline via tz wrote: It appears to me that where on network you access data.iana.org <https://data.iana.org/time-zones/releases/><https://data.iana.org/time-zones/releases/%3e> from might make a difference in what content you get at a given time. Are there mirrored servers that perhaps get delayed updates? I have an automated process that looks for new tzdata releases by periodically checking https://data.iana.org/time-zones/releases/ <https://data.iana.org/time-zones/releases/><https://data.iana.org/time-zones/releases/%3e>. I had received the mailing from this list that 2024b had been released, but had no sign from my automated process that the update had occurred. My polling interval is 30 minutes, so detection should be fairly fast. If used my own web browser at home to access https://data.iana.org/time-zones/releases/ <https://data.iana.org/time-zones/releases/><https://data.iana.org/time-zones/releases/%3e> I could see 2024b. Oddly enough, however, when I SSHed the remote server where my process runs, and used curl at the command line to check https://data.iana.org/time-zones/releases/ <https://data.iana.org/time-zones/releases/><https://data.iana.org/time-zones/releases/%3e>, 2024b wasn’t there yet. I wondered if my server had simply cached old data, but after a reboot of the server the same stale data came through without 2024b being shown as available. It was at least a full day, maybe more, before 2024b was visible from that remote server. Is this perhaps deliberate, to prevent too many people trying to access updated timezone data all at once? More likely that server or org network access is front-ended by a proxy server that has to scan everything before it caches it and/or allows anyone to access it. -- Take care. Thanks, Brian Inglis Calgary, Alberta, Canada La perfection est atteinte Perfection is achieved non pas lorsqu'il n'y a plus rien à ajouter not when there is no more to add mais lorsqu'il n'y a plus rien à retirer but when there is no more to cut -- Antoine de Saint-Exupéry
participants (3)
-
brian.inglis＠systematicsw.ab.ca -
James Mitchell -
Kerry Shetline