Error from Topaz fullBackupTo:
Locked
12
12
Error from Topaz fullBackupTo:
 
|
Ciao, i have a deployment system 3.1.0.6 I have some GemTools login error ....... The first think to do is a Backup from Topaz When i do it the system answer:
Some considerations? Thanks, Dario _______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Error from Topaz fullBackupTo:
|
|
Dario, could it be that you had a tODE / GemTools / Jade session opened while you tried the backup? You may also inspect: System gemCacheStatisticsForSessionId: 14 And figure out which is that gem that prevented the backup. Cheers, On Wed, Mar 9, 2016 at 5:10 PM, Trussardi Dario Romano via Glass <[hidden email]> wrote:
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Error from Topaz fullBackupTo:
|
|
Thanks Mariano,
the last time a do login from Gemtools the GemTools go in loop and i kill the relative process. After the gemstone_status report:
With this the Topaz login don't works. After i do the command: 1) stopNetldi gestionale_ldi 2) startNetldi gestionale The Topaz login now go but the fullBackup report the error as in my first e-mail.
In Topaz when i do: topaz 1>run System gemCacheStatisticsForSessionId: 14 % anArray Dario
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Error from Topaz fullBackupTo:
|
|
In reply to this post by GLASS mailing list
Dario,
This error mentions that the system is waiting for voting to complete ... presumably you ran an mfc shortly before starting the backup and one or more of the gems in the system hadn't had a chance to vote ... if you have some seaside gems sitting around the voting logic will cause those gems to get sent a SIGABORT and they will vote ... the seaside gems typically wake up every 30 seconds while to "check for a SIGABORT" so if you see this happen wait a minute and try again ... If you get the same error again, then as Mariano mentions, there is the possibility that you might have another topaz/tode/gemtools session alive and these guys typically don't "poll for SIGABORTS" ... Dale On 03/09/2016 12:10 PM, Trussardi Dario
Romano via Glass wrote:
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Error from Topaz fullBackupTo:
|
|
Dale,
The last fullBackup is date to yesterday...... i can't lose the today transactions. I can do the command: gemstone_stop and after restart with: gemstone_start or is too risky and it is better to make a fullBackup before ?
topaz 1>print System gemCacheStatisticsForSessionId: 14 % It's significant ? Thanks, Dario P.S. I have a Cron Backup procedure ( see the attachment ) running at 22:00 it works, create the file: gestionale_backup_20160309_220001.dbf
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
startGemStoneBackup.sh (3K)
Gemtools login problematic
|
|
Dale, I resubmit the fullBackup from Topaz, now it works fine. Now i I proceed with more tranquility..... Thanks. Now after do the backup i can work to the real problematic: the GemTools login don't works. I have the server based on Ubuntu system and all worked fine ( the Gemtools login worked fine [ i work into local network ] ) until the network man responsible change the router configurations. The public ip changed. In this server i run the Daemontools service. But now when i do the GemTools login the image go in loop [ i work from local network ] and after the gemstone_status report:
I restart the Ubuntu server but the problematic remain. The only thing i noticed is relative to the ssh login time ( 10 seconds ) from local connection, ( it's long time ). I tested the same Gemstone repository ( a copy of the same extent0.dbf ) on another local server and the Gemtools login works fine. The ssh login time relative to this server is 1 second. This server don't run the Daemontools service. Some considerations, idea about this problematic ? Thanks, Dario _______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Gemtools login problematic
|
|
Dario,
If you have tODE installed, I would like you try the `testLogin` command, since it gives me a nice report about what might be going wrong ... If you try to use port forwarding for GemTools[1], we will eliminate a couple of variables.. Now I notice that in your gslist, that the netldi is reported as `frozen` and that could be an issue. If you could send us a copy of the netldi log file we might be able to understand what might be going on ... but I don't think a frozen netldi is a good thing ... Also I'd like to see a copy of the settings you are using to log into your stone from GemTools ... Ip address changes could affect how GemTools host resolves ip addresses or how the stone resolves the host name for your GemTools host ... I would like a bit mor edetailo about which hosts had their ip addresses changed ... Dale [1] https://github.com/GsDevKit/GsDevKit_home/blob/master/docs/portForwardingForRemoteLogins.md On 03/09/2016 02:53 PM, Trussardi Dario
Romano via Glass wrote:
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Error from Topaz fullBackupTo:
|
|
In reply to this post by GLASS mailing list
On 03/09/2016 01:16 PM, Trussardi Dario
Romano via Glass wrote:
Regarding risk of losing data on system restart ... you should already be keeping copies of your backup files and tranlogs on a separate system, so that if you lose a disk, you have fairly recent copies of both ... if you are concerned about losing as little data as possible, you might think about copy tranlogs to a separate machine several times a day ... Anyway, you don't need to do a full backup right before cycling a server (it doesn't hurt to do so), but you might want to make sure that you've copied the tranlogs to your backup machine before cycling a server on the off chance that a disk fails on startup ... then you will be able to minimize the data that's lost if the system does crash ... If you are going to cycle a server, you might shut down the network or just shutdown the stone useing daemontools, then copy tranlogs, then cycle the machine ... that way you won't lose any tranlog data in the case of trouble restarting ... Dale _______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Gemtools login problematic
|
|
In reply to this post by GLASS mailing list
Dale,
ssh -X 192.168.1.10 -l scandella -L 57858:localhost:57858 In the hosts file on the macBook i have defined: 192.168.1.10 scandella ftp.scandella The system after 16 seconds answer:
After i open the tODE shell gestionale. The system answer in 1 second. The tODE commands works fine.
It's strange ? But the > scandella@scandella:/opt/oodb/gsDevKitHome/gemstone/stones/gestionale/product/bin$
./gslist -x report:
Thanks for considerations, Dario
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Gemtools login problematic
|
|
Dale,
The image go into loop. I kill it. Now the gslist -x report : The gemstone_status reports:
Now the tODE testLogin answer: Error running testLogin:
Dario
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Gemtools login problematic
|
|
Dario, you pasted the OGCustomSessionDescription from GemTools. Maybe you could also paste us the description file (GsDevKit_home/sys/local/sessions/yourStoneFile) of such a stone to see if there is a difference in the connection settings. Cheers, On Thu, Mar 10, 2016 at 8:50 AM, Trussardi Dario Romano via Glass <[hidden email]> wrote:
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
Re: Gemtools login problematic
|
|
Ciao Mariano,
There is no match with your directory paths indications. Thanks, Dario
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
|
|
In reply to this post by GLASS mailing list
Ciao, some time ago ( 10 marzo 2016 ) Dale wrote :
Now i have a small server ( with disk under RAID 5 software and with UPS ) where run Gemstone 3.1.0.6 The UPS manage the shutdown system ( when power go down ) with the procedure: gemstone_batchstop
When Dale wrote: " if you are concerned about losing as little data as possible, " i'm scared. My system is very small but if i think to complex system, i can't not losing data, not any transaction. I have not experienced, and maybe I worry too much, but i 'need' a system where losing data it is a very, very remote possibility. In the last year i don't have any problem on the server ( run 24x7 ) but i need to clarify the situation. Some experience? considerations? Thanks, Dario
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
|
|
In reply to this post by GLASS mailing list
Ciao, some time ago ( 10 marzo 2016 ) Dale wrote :
OK, i working in this direction. I have some doubts: A) I need a external script for create the new tranlog every hour with the System class >> startNewLog. In general i have some methods to perform at regular time and data but i don't understand if i can schedule it directly into the gemstone, or if i need external scripts call by Crontab? B) On LAN i have a Windows system for perform ftp copy from the Gemstone server to the backup system. ( i use the Cobian program ) Now i think to setup the incremental ftp copy of the /tranlogs server directory. But how does the ftp command by copying the online tranlog ? Can create problematic on the server? C) Link to my last email : [Glass] Integrity of the system I read about Warm and Hot Standbys Is not possible have the " online copy " of the current tranlog on the backup system ? Where for online copy i think the copy at the last transaction ( update at the last minute?! ) on the server. If this would be possible, when the server crash, on the backup system i found the tranlog update to the latest transactions. Is not bad? Am I asking too much? Thanks for any considerations, Dario
_______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
|
|
We keep a warm recovery DB for our production GS system on a geographically separate system. Nightly we restore the recovery DB from the latest prod backup, then as we switch tranlogs (script via cron) we have scripts that will package it up, send it over to the restore system, kick off scripts on that side that will prepare the tranlog and apply it to the warm recovery system. Both sides are on AIX and we use rcp to move the files. You cannot copy active tranlogs nor extents (we use raw) so requires the backup written to disk and exporting tranlogs to filesystems to be able to move them over. We are on a fairly old version so newer versions may be more robust. On Thu, Jul 28, 2016 at 1:00 PM, Trussardi Dario Romano via Glass <[hidden email]> wrote:
-B- "Think of how stupid the average person is, and realize half of 'em are stupider than that" - George Carlin _______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
|
|
In reply to this post by GLASS mailing list
Seems as if my mail went directly to Dario, so just copying it here if
anyone is interested. ---------- Forwarded message ---------- From: Otto Behrens <[hidden email]> Date: Fri, Jul 29, 2016 at 8:26 AM Subject: Re: [Glass] Backup procedure To: Trussardi Dario Romano <[hidden email]> Hi Dario, > Regarding risk of losing data on system restart ... you should already be > keeping copies of your backup files and tranlogs on a separate system, so > that if you lose a disk, you have fairly recent copies of both ... if you > are concerned about losing as little data as possible, you might think about > copy tranlogs to a separate machine several times a day ... Every day, we run SystemRepository fullBackupCompressedTo:, and then scp the file to a backup server > A) I need a external script for create the new tranlog every hour with > the System class >> startNewLog. > In general i have some methods to perform at regular time and data but i > don't understand if i can schedule it directly into the gemstone, > > or if i need external scripts call by Crontab? You can do this from cron. But GemStone has settings to control when to write tranlogs, if you require more frequent logs. We do not do this, we use the hot standby. See below. > B) On LAN i have a Windows system for perform ftp copy from the Gemstone > server to the backup system. > ( i use the Cobian program ) > > Now i think to setup the incremental ftp copy of the /tranlogs server > directory. > > But how does the ftp command by copying the online tranlog ? > > Can create problematic on the server? If you do startNewLog from your cron script, it tells you which log was created. You can then safely copy the previous logs away. Again, we do not do this, we rely on the standby. > C) Link to my last email : [Glass] Integrity of the system > > I read about Warm and Hot Standbys Yes, we use this. LogSender on production and LogReceiver on standby. > Is not possible have the " online copy " of the current tranlog on the > backup system ? The hot standby system does that. As log entries are written on your production system, they are copied across to your standby system (into the same tranlog file). > Where for online copy i think the copy at the last transaction ( update at > the last minute?! ) on the server. Yes, the logsender / logreceiver updates almost as it happens. > If this would be possible, when the server crash, on the backup system i > found the tranlog update to the latest transactions. > Is not bad? You can do this, yes. We have recovered from the hot standby before. > Am I asking too much? Nope, it is there. Managing the eco-system of production and hot standby is a bit of work. This is to ensure that it actually *is* running. For us, it is not enough to just start the logsender and logreceiver. We want to be sure that it is working. So, on our standby server, we parse the log output and log for signs of recent syncing activity. This is a ruby script. So, if you'd like to have a look at the scripts, let me know. Also, we restore the overnight backup and restart the syncing process on the standby machine every day. Strictly speaking, you don't need to do this every day (although it's advisable to do on some frequency, I think). We do this to keep the extent small and to reduce the number of transaction logs we need. Cheers Otto _______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
|
|
In reply to this post by GLASS mailing list
We used to do this but switched over to the logsender / logreceiver
(hot standby) some time ago. It did make management of copying of tranlogs over to the standby system easier. I recommend you use this if you can; it is robust. On Thu, Jul 28, 2016 at 7:55 PM, Bob Tucker via Glass <[hidden email]> wrote: > We keep a warm recovery DB for our production GS system on a geographically > separate system. Nightly we restore the recovery DB from the latest prod > backup, then as we switch tranlogs (script via cron) we have scripts that > will package it up, send it over to the restore system, kick off scripts on > that side that will prepare the tranlog and apply it to the warm recovery > system. Both sides are on AIX and we use rcp to move the files. You cannot > copy active tranlogs nor extents (we use raw) so requires the backup written > to disk and exporting tranlogs to filesystems to be able to move them over. > We are on a fairly old version so newer versions may be more robust. > > On Thu, Jul 28, 2016 at 1:00 PM, Trussardi Dario Romano via Glass > <[hidden email]> wrote: >> >> Ciao, >> >> some time ago ( 10 marzo 2016 ) Dale wrote : >> >> Regarding risk of losing data on system restart ... you should already be >> keeping copies of your backup files and tranlogs on a separate system, so >> that if you lose a disk, you have fairly recent copies of both ... if you >> are concerned about losing as little data as possible, you might think about >> copy tranlogs to a separate machine several times a day ... >> >> >> OK, i working in this direction. >> >> I have some doubts: >> >> A) I need a external script for create the new tranlog every hour with >> the System class >> startNewLog. >> In general i have some methods to perform at regular time and data but i >> don't understand if i can schedule it directly into the gemstone, >> >> or if i need external scripts call by Crontab? >> >> B) On LAN i have a Windows system for perform ftp copy from the Gemstone >> server to the backup system. >> ( i use the Cobian program ) >> >> Now i think to setup the incremental ftp copy of the /tranlogs server >> directory. >> >> But how does the ftp command by copying the online tranlog ? >> >> Can create problematic on the server? >> >> C) Link to my last email : [Glass] Integrity of the system >> >> I read about Warm and Hot Standbys >> >> Is not possible have the " online copy " of the current tranlog on the >> backup system ? >> >> Where for online copy i think the copy at the last transaction ( update at >> the last minute?! ) on the server. >> >> If this would be possible, when the server crash, on the backup system i >> found the tranlog update to the latest transactions. >> Is not bad? >> >> Am I asking too much? >> >> Thanks for any considerations, >> >> Dario >> >> >> >> _______________________________________________ >> Glass mailing list >> [hidden email] >> http://lists.gemtalksystems.com/mailman/listinfo/glass >> > > > > -- > -B- > > "Think of how stupid the average person is, and realize half of 'em are > stupider than that" > - George Carlin > > _______________________________________________ > Glass mailing list > [hidden email] > http://lists.gemtalksystems.com/mailman/listinfo/glass > Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
|
|
Thanks for your considerations!!!
From my point of view: > We used to do this but switched over to the logsender / logreceiver > (hot standby) some time ago. It did make management of copying of > tranlogs over to the standby system easier. I recommend you use this > if you can; it is robust. OK. This is the top of backup procedure. And if i right understand i can switch from master to slave in real time ( some minutes? ). But it required some hardware and .... configuration..... I think to use it to very 24x7 maximum guarantee system and continuity of service. My goal is to have a small system which can break, but with support to restore at the last transaction. In case of the system crash, i don't need to restart in few minutes, but to restore at the last transaction. The customer knows that the system can crash, and in this case the system can stay down for some hours, but when the system restart it's update at the last transaction. Restore the system to the last tranlog backup is not sufficient. As the customer can realign the system if the last backup tranlog is 30 minutes old? Where i found the last 30 minutes transaction if the server is dead? I think having an intermediate solution could resolve many situations. Having the ability to have a online copy of the active tranlog on the backup system it's a very good solution. Is possible use the logsender / logreceiver only to replicate the current logs on the backup system ? With this support and the last production backup it's possible restore the system without losing any transactions. But these are my considerations..... Thanks, Dario > On Thu, Jul 28, 2016 at 7:55 PM, Bob Tucker via Glass > <[hidden email]> wrote: >> We keep a warm recovery DB for our production GS system on a geographically >> separate system. Nightly we restore the recovery DB from the latest prod >> backup, then as we switch tranlogs (script via cron) we have scripts that >> will package it up, send it over to the restore system, kick off scripts on >> that side that will prepare the tranlog and apply it to the warm recovery >> system. Both sides are on AIX and we use rcp to move the files. You cannot >> copy active tranlogs nor extents (we use raw) so requires the backup written >> to disk and exporting tranlogs to filesystems to be able to move them over. >> We are on a fairly old version so newer versions may be more robust. >> >> On Thu, Jul 28, 2016 at 1:00 PM, Trussardi Dario Romano via Glass >> <[hidden email]> wrote: >>> >>> Ciao, >>> >>> some time ago ( 10 marzo 2016 ) Dale wrote : >>> >>> Regarding risk of losing data on system restart ... you should already be >>> keeping copies of your backup files and tranlogs on a separate system, so >>> that if you lose a disk, you have fairly recent copies of both ... if you >>> are concerned about losing as little data as possible, you might think about >>> copy tranlogs to a separate machine several times a day ... >>> >>> >>> OK, i working in this direction. >>> >>> I have some doubts: >>> >>> A) I need a external script for create the new tranlog every hour with >>> the System class >> startNewLog. >>> In general i have some methods to perform at regular time and data but i >>> don't understand if i can schedule it directly into the gemstone, >>> >>> or if i need external scripts call by Crontab? >>> >>> B) On LAN i have a Windows system for perform ftp copy from the Gemstone >>> server to the backup system. >>> ( i use the Cobian program ) >>> >>> Now i think to setup the incremental ftp copy of the /tranlogs server >>> directory. >>> >>> But how does the ftp command by copying the online tranlog ? >>> >>> Can create problematic on the server? >>> >>> C) Link to my last email : [Glass] Integrity of the system >>> >>> I read about Warm and Hot Standbys >>> >>> Is not possible have the " online copy " of the current tranlog on the >>> backup system ? >>> >>> Where for online copy i think the copy at the last transaction ( update at >>> the last minute?! ) on the server. >>> >>> If this would be possible, when the server crash, on the backup system i >>> found the tranlog update to the latest transactions. >>> Is not bad? >>> >>> Am I asking too much? >>> >>> Thanks for any considerations, >>> >>> Dario >>> >>> >>> >>> _______________________________________________ >>> Glass mailing list >>> [hidden email] >>> http://lists.gemtalksystems.com/mailman/listinfo/glass >>> >> >> >> >> -- >> -B- >> >> "Think of how stupid the average person is, and realize half of 'em are >> stupider than that" >> - George Carlin >> >> _______________________________________________ >> Glass mailing list >> [hidden email] >> http://lists.gemtalksystems.com/mailman/listinfo/glass >> > _______________________________________________ > Glass mailing list > [hidden email] > http://lists.gemtalksystems.com/mailman/listinfo/glass _______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
|
|
> And if i right understand i can switch from master to slave in real time ( some minutes? ).
When you need to activate your slave, you 1. stop the logreceiver 2. SystemRepository commitRestore This is really quick (a few seconds) 3. Switch over your web server to talk to the slave. Here you may have to think a bit how you want to do it. If your web server is on master, you may have to do a DNS change (if the master machine is dead). If it is alive, you can route traffic to the slave using the firewall or some other way. > But it required some hardware and .... configuration..... > > I think to use it to very 24x7 maximum guarantee system and continuity of service. Yes, I think it is a decent solution. There is no absolute guarantee though. You have to be sure that the logsender / receiver have caught up all the log entries and you may have switched over on a point where you lost a last transaction. > Restore the system to the last tranlog backup is not sufficient. > > As the customer can realign the system if the last backup tranlog is 30 minutes old? I don't really get what you're saying here. If you have the tranlogs on master, then you can replay until the last transaction, when your system comes back up. If your tranlogs are lost on master but you restart with the existing extent(s), you may have lost transactions. If you have a hot standby in place, you are as up to date as you can be if you lose tranlogs on master. If not losing tranlogs on master is really critical, I'll consider RAID and what other "safe" disk options you can get to store your tranlogs on. Gemstone also has a replicate tranlog directory which you can mount on a remote machine or something else. > Where i found the last 30 minutes transaction if the server is dead? On the hot standby > I think having an intermediate solution could resolve many situations. > > Having the ability to have a online copy of the active tranlog on the backup system it's a very good solution. > > Is possible use the logsender / logreceiver only to replicate the current logs on the backup system ? If I did run a hot standby, I do not worry about replicating tranlogs because this is what the logsender / receiver does > With this support and the last production backup it's possible restore the system without losing any transactions. > > But these are my considerations..... > > Thanks, > > Dario > > >> On Thu, Jul 28, 2016 at 7:55 PM, Bob Tucker via Glass >> <[hidden email]> wrote: >>> We keep a warm recovery DB for our production GS system on a geographically >>> separate system. Nightly we restore the recovery DB from the latest prod >>> backup, then as we switch tranlogs (script via cron) we have scripts that >>> will package it up, send it over to the restore system, kick off scripts on >>> that side that will prepare the tranlog and apply it to the warm recovery >>> system. Both sides are on AIX and we use rcp to move the files. You cannot >>> copy active tranlogs nor extents (we use raw) so requires the backup written >>> to disk and exporting tranlogs to filesystems to be able to move them over. >>> We are on a fairly old version so newer versions may be more robust. >>> >>> On Thu, Jul 28, 2016 at 1:00 PM, Trussardi Dario Romano via Glass >>> <[hidden email]> wrote: >>>> >>>> Ciao, >>>> >>>> some time ago ( 10 marzo 2016 ) Dale wrote : >>>> >>>> Regarding risk of losing data on system restart ... you should already be >>>> keeping copies of your backup files and tranlogs on a separate system, so >>>> that if you lose a disk, you have fairly recent copies of both ... if you >>>> are concerned about losing as little data as possible, you might think about >>>> copy tranlogs to a separate machine several times a day ... >>>> >>>> >>>> OK, i working in this direction. >>>> >>>> I have some doubts: >>>> >>>> A) I need a external script for create the new tranlog every hour with >>>> the System class >> startNewLog. >>>> In general i have some methods to perform at regular time and data but i >>>> don't understand if i can schedule it directly into the gemstone, >>>> >>>> or if i need external scripts call by Crontab? >>>> >>>> B) On LAN i have a Windows system for perform ftp copy from the Gemstone >>>> server to the backup system. >>>> ( i use the Cobian program ) >>>> >>>> Now i think to setup the incremental ftp copy of the /tranlogs server >>>> directory. >>>> >>>> But how does the ftp command by copying the online tranlog ? >>>> >>>> Can create problematic on the server? >>>> >>>> C) Link to my last email : [Glass] Integrity of the system >>>> >>>> I read about Warm and Hot Standbys >>>> >>>> Is not possible have the " online copy " of the current tranlog on the >>>> backup system ? >>>> >>>> Where for online copy i think the copy at the last transaction ( update at >>>> the last minute?! ) on the server. >>>> >>>> If this would be possible, when the server crash, on the backup system i >>>> found the tranlog update to the latest transactions. >>>> Is not bad? >>>> >>>> Am I asking too much? >>>> >>>> Thanks for any considerations, >>>> >>>> Dario >>>> >>>> >>>> >>>> _______________________________________________ >>>> Glass mailing list >>>> [hidden email] >>>> http://lists.gemtalksystems.com/mailman/listinfo/glass >>>> >>> >>> >>> >>> -- >>> -B- >>> >>> "Think of how stupid the average person is, and realize half of 'em are >>> stupider than that" >>> - George Carlin >>> >>> _______________________________________________ >>> Glass mailing list >>> [hidden email] >>> http://lists.gemtalksystems.com/mailman/listinfo/glass >>> >> _______________________________________________ >> Glass mailing list >> [hidden email] >> http://lists.gemtalksystems.com/mailman/listinfo/glass > > _______________________________________________ > Glass mailing list > [hidden email] > http://lists.gemtalksystems.com/mailman/listinfo/glass Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
|
|
Ciao,
very thanks.... >> And if i right understand i can switch from master to slave in real time ( some minutes? ). > > When you need to activate your slave, you > 1. stop the logreceiver > 2. SystemRepository commitRestore > > This is really quick (a few seconds) > > 3. Switch over your web server to talk to the slave. Here you may have > to think a bit how you want to do it. If your web server is on master, > you may have to do a DNS change (if the master machine is dead). If it > is alive, you can route traffic to the slave using the firewall or > some other way. Very well. > >> But it required some hardware and .... configuration..... >> >> I think to use it to very 24x7 maximum guarantee system and continuity of service. > > Yes, I think it is a decent solution. There is no absolute guarantee > though. You have to be sure that the logsender / receiver have caught > up all the log entries and you may have switched over on a point where > you lost a last transaction. > >> Restore the system to the last tranlog backup is not sufficient. >> >> As the customer can realign the system if the last backup tranlog is 30 minutes old? > > I don't really get what you're saying here. Ok, excuse. I'm thinking about system without any standby system. But a system where i create a new tranlogs every 30 minutes and i save the old tranlogs on the backup system every 30 minutes. In this case on the backup system i have the last repository backup ( do every night ) and the relative old tranlogs ( create in sequence every 30 minutes ) but i don't have any "copy" of the current tranlogs. In this case if the master system go down i lose the last online tranlogs ( with the transactions relative to the current 30 minutes ) > If you have the tranlogs on master, then you can replay until the last > transaction, when your system comes back up. > If your tranlogs are lost on master but you restart with the existing > extent(s), you may have lost transactions. > If you have a hot standby in place, you are as up to date as you can > be if you lose tranlogs on master. > > If not losing tranlogs on master is really critical, I'll consider > RAID and what other "safe" disk options you can get to store your > tranlogs on. Gemstone also has a replicate tranlog directory which you > can mount on a remote machine or something else. I don't know this options. It's a possible solutions ! ? The Gemstone replicate the online tranlogs in this directory on 'external' system ? If yes, i have a "copy" of the online tranlogs and i can restore from backup system to the last transaction. I don't find reference to this options in the System Administration Guide 3.1. > >> Where i found the last 30 minutes transaction if the server is dead? > > On the hot standby I write these considerations, thinking to a system without standby. > >> I think having an intermediate solution could resolve many situations. >> >> Having the ability to have a online copy of the active tranlog on the backup system it's a very good solution. >> >> Is possible use the logsender / logreceiver only to replicate the current logs on the backup system ? I need only to have the "copy" of the online tranlogs. > > If I did run a hot standby, I do not worry about replicating tranlogs > because this is what the logsender / receiver does > >> With this support and the last production backup it's possible restore the system without losing any transactions. >> >> But these are my considerations..... >> Thanks, CIAO, Dario _______________________________________________ Glass mailing list [hidden email] http://lists.gemtalksystems.com/mailman/listinfo/glass |
«
Return to GLASS
|
1 view|%1 views
| Free forum by Nabble | Edit this page |