F~u~c~k i~t!
Would you like to react to this message? Create an account in a few clicks or log in to continue.
Search
 
 

Display results as :
 


Rechercher Advanced Search

Top posting users this month
No user

GeoMap

ClustrMaps server hard disk damaged, user data may loss permanently!

Go down

ClustrMaps server hard disk damaged, user data may loss permanently! Empty ClustrMaps server hard disk damaged, user data may loss permanently!

Post  Admin Tue Mar 31, 2015 10:02 pm

The company declared that the user data is still safe on the "dead" server's disk, however they made little progress on retrieve user data from them. For an IT company, this should never happened, they should really be careful next time.


If your map was stored on our “www4″ server (you can tell because you can see “www4.clustrmaps.com” or ” ’server’ : ‘4′ ” in the HTML where you embedded the ClustrMaps thumbnail map), then some major changes have taken place in the last few days (March 23-25th 2015), and we recommend that you carefully read all the items below.

Your data is safely stored*, but regrettably server www4 has now been declared ‘dead’.
*(Safe storage of data is true for the overwhelming majority of users – a very small proportion of data was lost during the series of crashes that plagued us over the last few weeks, and we will know more about this over the coming weeks as we provide retrieval details – see next item).
We will provide details about how to retrieve items like your most recent visitor totals and more in a later blog posting or special page, to which we will provide a prominent link (here and in other postings) when it is ready.
As of March 24th 2015, Your originally-registered URL will not be recognised on the normal Admin login page at [You must be registered and logged in to see this link.] (but see next item).
We recommend that you register your URL again just like a new user would register, on our home page at [You must be registered and logged in to see this link.] or directly on [You must be registered and logged in to see this link.] This will mean that a fresh count will be started at 0, but once you can retrieve your earlier data (as mentioned above), then you will be able to provide a ‘footnote’ or ‘caveat’ near your new map that describes the earlier visitor data.
Paying (ClustrMaps+) accounts: If you are a paying (ClustrMaps+) user on the affected server you have two options (re-register vs refund), as follows -
You can re-register as suggested above, but it may take us a few days to get you properly upgraded to ClustrMaps+ status for the re-registered map, since this cannot be automated right now; if you are in a particular hurry, you can contact us at [You must be registered and logged in to see this link.] and be sure to mention ‘upgrade’ and tell us your registered ClustrMaps+ URL (or multiple URLs) and approximately when your subscription(s) started. Please be sure to re-read items 2 and 4 above to make sure you understand that your new maps will start from 0 and that it will take a while before you can get access to any previous data.
Refund for paid accounts: If you do not want to re-register your URL as suggested above but would instead prefer a pro rata (proportional) refund covering the outage period from March 9th 2015 to the end of your subscribed period, then please contact us at [You must be registered and logged in to see this link.] and be sure to mention ‘refund’ and tell us your registered ClustrMaps+ URL (or multiple URLs) and approximately when your subscription(s) started.
Could such a terrible crash happen again on another server? Sadly, the honest answer is “Yes”. We take great precautions, and have great backup systems, but they are not infallible, especially for a small company like ClustrMaps Ltd.
As you can appreciate, we are devastated by this unexpected loss, and the impact it has on thousands of our loyal users. We have worked very hard over the last few weeks to bring www4 back to life, but it appears that the complexity of the ‘cascading’ problem defeated our best efforts and overwhelmed the resources we could allocate. We would have notified everyone sooner but it was not possible due to a combination of (a) we were optimistic about finding a solution, as we have always done in the past and (b) there are too many users with expired email addresses, so we have to resort to this blog posting and a link to it through the thumbnail map you see at the top of this post.

We are overwhelmed with emails, so unless there is a particular dilemma you need urgently to bring to our attention, we would be grateful if you only contact us concerning refunds, as mentioned in the numbered items above.

We’d like to express our apologies once again and thank the many users who sent us their wishes and support.

[Final update] Map/count problems on www4 March 12, 2015
Posted by ClustrMaps Team in : Uncategorized , comments closed
[FINAL UPDATE, 25-March-2015 01:00 GMT: This post has now been superseded by a later and more important news item concerning the fate of www4. Please see [You must be registered and logged in to see this link.] or scroll 'higher up' on the main blog.clustrmaps.com ]

[UPDATE 10, 23-March-2015 22:50 GMT: Not much progress beyond UPDATE 9 at this point (see that update below if you are a paying user) - we note that users' patience is starting to run out, so we're urgently looking into other options to try to restore the service or at least arrive at a more definitive statement of its fate. Please bear with us... thanks!]

[UPDATE 9, 21-March-2015 11:00 GMT: Refunds? It looks like this terrible interruption of service is going to be continuing for a while. We are devastated by this outage, but it has grown in complexity over the past few days, just for this specific server www4.clustrmaps.com. If you are a paying ClustrMaps+ user, and you would like a pro rata (proportional) refund, please contact us at [You must be registered and logged in to see this link.] and please tell us your registered ClustrMaps+ URL and approximately when your subscription started. Many thanks again to all.]

[UPDATE 9, 19-March-2015 00:20 GMT: Slow progress - a major data transfer and recovery operation is underway. Our data sets are massive, and require a significant transfer and recovery operation due to the multiple faults that happened on this server. This may take a few more days, I'm afraid - there were multiple underlying problems that need to be resolved, and we will provide further updates as soon as we have some more tangible progress to report.]

[UPDATE 8, 18-March-2015 13:00 GMT: New disks failing - deeper investigation underway. This is very dis-heartening for us, believe me... but staying optimistic.]

[UPDATE 7, 18-March-2015 00:45 GMT: Server replaced, maps running fine. In fact they have now been running for the past 8+ hours, since about 16:00 GMT 17th March, but we wanted to let it run for a while to confirm that all was OK. The map updates with the latest counts should start appearing during the next two days.]

[UPDATE 6, 17-March-2015 12:00 noon GMT: Chassis swap completed overnight... but... there's still a lot of work to do re-installing and verifying all the services... we're hoping all will be OK by tonight, but you can see from the long ongoing saga in these updates that it would be unwise to promise anything at this point...]

[UPDATE 5, 17-March-2015 00:15 GMT: Better news -- after the previous update (immediately below), things were looking like they might take a few more days, but now we're talking about hours rather than days, as a (simpler) 'chassis swap' is underway... new motherboards and controllers basically, and the shiny new disks already loaded up with all our software and data, so we're optimistic again. It's been a real roller-coaster ride, and the support of our thousands of fans and users is greatly appreciated!]

[UPDATE 4, 16-March-2015 22:45 GMT: Bad news -- well, we thought we had it solved with a double disk replacement and rebuild! Alas, there are ongoing issues with the (old) server itself, and this probably needs replacing too, which is going to mean an outage of several more dayshours, we suspect. This affects a very large number of users, so we are investigating some urgent measures such as moving everyone over to www3, but in a way that does not require you to change the HTML on your page (by redirecting traffic accordingly)... we'll let you know as soon as we have more news. We're really extremely sorry about this outage, and appreciate your sticking with us through this difficult period.]

[UPDATE 3, 16-March-2015 00:15 GMT: The disk array for this server (www4) has now been completely replaced and rebuilt, and is now running reliably again. Maps are working fully and counting properly, but you may still see the latest date for the statistics as '13th March 2015': this will be brought up to date with the latest data over the next few days. Please stay tuned.... ]

[UPDATE 2, 14-March-2015 01:00 GMT: New problems, it seems there are still some 'up and down' issues, so we will have to initiate some deeper investigations regarding the underlying causes. Please stay tuned.... ]

[UPDATE 1, 12-March-2015 11:30 GMT: This has now been (temporarily) fixed, the server is up and running, large maps on www4 will show 'Statistics updated 11-Mar-2015 (and will catch up successively overnight tonight).]

[ORIGINAL POST BELOW]
If you are experiencing unusual delays with ‘Statistics updated’ (not changed since 7th March), ‘recent visits’ or even ‘disappearing thumbnail maps’, and your map’s HTML code begins with www4.clustrmaps.com, please note the following: multiple ‘outage’ periods on www4.clustrmaps.com from 9th March – 12th March (5 hours on 9th March, much shorter periods on successive days) also upset the count refresh, so for some users it means the ’statistics updated’ appears to be stuck and the ‘recent visitors’ (yellow dots) had to restart from scratch.

The good news is that our log file is continuously logging all traffic*, so your count will be updated as soon as we can get to the bottom of the current logjam (which, for the technically-minded is due to a database synchronization issue across our multiple servers). We will be posting updates right here on this posting, so please check back from time to time to see what the current situation. Many thanks for your patience.
——
*logging of traffic is only possible when your thumbnail map is displayed

Admin
Admin

帖子数 : 5
金币(Coin) : 4254
注册日期 : 2012-08-16

https://fuckit.longluntan.com

Back to top Go down

Back to top

- Similar topics

 
Permissions in this forum:
You cannot reply to topics in this forum