pdumigan's Posts

pdumigan's Posts

I'm trying to troubleshoot connectivity on a storage node.  I see I the documentation that you can set the debug level on a client, and processes.  Sometimes it says to set it to a number between... See more...
I'm trying to troubleshoot connectivity on a storage node.  I see I the documentation that you can set the debug level on a client, and processes.  Sometimes it says to set it to a number between 1-9 or 1-99. Are these progressive levels or does anything greater than 0 produce the same output?  If they are progressive, where can I find some guidance on the level to choose? Thanks.
Thanks!  So it looks like I am getting connection lost errors every six hours. The excel formula for converting to Eastern US (EDT) is ([sourceCell]/86400)+25569+(-4/24).  Then format that as ... See more...
Thanks!  So it looks like I am getting connection lost errors every six hours. The excel formula for converting to Eastern US (EDT) is ([sourceCell]/86400)+25569+(-4/24).  Then format that as a date/time.
I have a Networker 9 storage node that keeps reporting "errno = An existing connection was forcibly closed by the remote host" in the daemon.raw. However, after rendering the log, at the begin... See more...
I have a Networker 9 storage node that keeps reporting "errno = An existing connection was forcibly closed by the remote host" in the daemon.raw. However, after rendering the log, at the beginning of the line it says that it is "Unable to render the following message" so I can't determine the date and time of the error.  I see that the date/time is stored in some kind of serial number(1468928259) format in the RAW file. Is there a way to convert that number to a date/time so I can try to figure out what is killing the connection?
The new server will have more than enough storage to double that of the original.  I wonder if I could recreate the file devices using the same folder paths and copy over the device contents to t... See more...
The new server will have more than enough storage to double that of the original.  I wonder if I could recreate the file devices using the same folder paths and copy over the device contents to the new server?  That way, wouldn't the server not know the device/file media had been moved?
But if the new server will not be attached to the old server's devices, won't the restored bootstrap reference devices that no longer exist?
I am running two Networker servers. 1) Windows Server/Storage Node/NMC has 1 standalone LTO SAS tape drive for cloning, 6 'file' type devices on direct attached SAS hardware for active backups... See more...
I am running two Networker servers. 1) Windows Server/Storage Node/NMC has 1 standalone LTO SAS tape drive for cloning, 6 'file' type devices on direct attached SAS hardware for active backups, and 3 'file' type devices on direct attached SAS hardware for active staging. 2) Windows Storage Node has 1 direct attached SAS LTO tape library for active backups. Both servers are running NW 8.2.1.2. Server 1 and its attached storage hardware is ready for retirement.  I am purchasing a new server with direct attached storage to replace it.  The new server will run NW 8.2.1.2 as well. I ran the SoIVe program to find the steps for migrating to a new server with a new name.  That part is fairly clear.  What it doesn't tell me is how to get the 'file' devices, media, and savesets over to the new hardware.  Is there instructions for doing that or how can that be done?
How would I move data from it?  Can I stage back into the same pool but onto the other devices?
I have a drive/disk volume that I'm using in a staging pool to stage older save sets off of my primary disk backup pool.  The drive is one of four in the staging pool.  It is a stand-alone drive ... See more...
I have a drive/disk volume that I'm using in a staging pool to stage older save sets off of my primary disk backup pool.  The drive is one of four in the staging pool.  It is a stand-alone drive that I needed to use to expand the staging pool for successful staging.  The rest of the staging pool is on RAID volumes.  The drive is now starting to report some disk errors, so I would like to remove it from use.  I should be able to operate with just the other three now.  It has some active save sets staged on it right now but they are due to expire in the next couple of weeks. Is there a documented method for retiring this volume and disk from use?  Should I just mark it as read-only until all of the save sets expire?  If I mark it as read only will the expired save sets still be deleted or do they need to be?
While continuing to try to troubleshoot VSS errors on my local 7.6.2 backup server, I found in the Windows logs the the VSS errors are being generated in reference to the local file device drives... See more...
While continuing to try to troubleshoot VSS errors on my local 7.6.2 backup server, I found in the Windows logs the the VSS errors are being generated in reference to the local file device drives.  Each local file device is a directory on a dedicated volume (s:\LocalDiskFile01, t:\LocalDiskFile02).  Networker has placed nsr.dir files in these folders that reads "# Don't backup ssid files null: volume skip: *" However, the OS still wants to volumesnap the volume.  Does the null: volume need to be at the root of the volume rather than in the directory?
Carlos,  I believe you are mistaken.  The All saveset does indeed include DISASTER_RECOVERY and this is mentioned a number of times in the Admin Guide and other documentation.  To not include it ... See more...
Carlos,  I believe you are mistaken.  The All saveset does indeed include DISASTER_RECOVERY and this is mentioned a number of times in the Admin Guide and other documentation.  To not include it would need an explicit "skip".  However, that means we now need to block out time for additional disaster recovery backups in what are already too short backup time windows.  This probably would be disruptive to production.  The system should just understand not to try to VSS snaphot the same drive more than once at the same time.  Hopefully this has been or will be addressed in a 762 hotfix.  I need to review the fixes to be sure.  The review alone is a tedious task. Maarten, the OSSR_ save sets are all explained starting on page 600 of the admin guide.  I'm reluctant to apply the MS hotfix, since the error it is claimed to repair is not the error I'm receiving.  Most firewall session errors can be aleviated by setting a system variable NSR_KEEPALIVE_WAIT on the client.  it is the number of seconds between session pings to keep the connection open while the client is working.  We use 28 and it works well.  search for it in the release notes for details
I am running Networker server network edition 7.6.2. Occasionally, I would receive "VSS Backup or Restore is not in the correct state for the operation" errors in my savegroup completion repor... See more...
I am running Networker server network edition 7.6.2. Occasionally, I would receive "VSS Backup or Restore is not in the correct state for the operation" errors in my savegroup completion reports for various servers. Networker documentation says that this is a problem that can occur on Windows clients if the parallelism is set to greater than 4 in the client's global properties. I reduced all client parallelisms to 4 and had pratically eliminated the problem. My Networker server is a Windows 2008 SP2 (not R2) enterprise server that is also a storage node with 3 local disk devices and one local tape device. I backup the local server to the local server using the ALL saveset using a weekly FULL and daily INCR schedule. Upgrading to 7.6 has enabled DISASTER_RECOVERY in the ALL saveset for Windows 2008. Ever since that has happened, I am frequently getting the VSS errors for the server's local client backup of itself when doing FULL backups. Not always, but frequently. Experience tells me that this is a parallelism problem. It would make sense that VSS is struggling, especially if it is trying to snapshot both C: and OSSR_C at the same time. I tried changing the parallelism for the server's client profile to 4, but after the next backup, Networker has changed it back to 12. I'm not sure what to do next. Is there a requirement that the server needs parallelism at 12? Any other ideas?
I am running networker 7.6.1.3 and trying to install the hotfix 7.6.1.6 but the installer keeps going into maintenance mode rather than update mode. I'm not sure what to do next. The documentati... See more...
I am running networker 7.6.1.3 and trying to install the hotfix 7.6.1.6 but the installer keeps going into maintenance mode rather than update mode. I'm not sure what to do next. The documentation says I should see update mode from the installer.  Instead, maintenance mode gives me the option to change, repair, or remove the software.  None of these seem appropriate.  When I choose one of them, the installer does not seem to be aware of what is currently installed. I remember in older versions to install a hotfix I needed to uninstall, choosing to keep the database, and reinstall the updated version.  Is that still the case in 7.6.1?  The documentation is confusing. I'm running Networker server network, storage node, management console, and client on this Windows 2008 Ent x64 server.
Just what I needed to know.  This definitely helps, thanks! The VMs are clients themselves at their OS level on the Networker server as well as being replicated at the SAN.  Support for the C... See more...
Just what I needed to know.  This definitely helps, thanks! The VMs are clients themselves at their OS level on the Networker server as well as being replicated at the SAN.  Support for the CSV is not an immediate need if the cluster node will still backup successfully.  It would be a nice feature in the future so I will submit the RFE.  In the meantime, I'm hoping I will be able to configure a SKIP of the CSV volumes ("C:\ClusterStorage\*" or similar) on the client resource to alleviate the errors  in the backup report.
My Networker server and storage installation is a stand alone installation.  My questions are not about running the server in a cluster.  Only about having cluster components as clients. I am ... See more...
My Networker server and storage installation is a stand alone installation.  My questions are not about running the server in a cluster.  Only about having cluster components as clients. I am in the process of building a Microsoft Windows 2008 R2 Hyper-V cluster.  To take advantage of many of the new features in R2, implementing Cluster Shared Volumes is a requirement.  I would like to add the physical cluster nodes as clients on my Networker server.  I read in the release notes for 7.5.2 that Cluster Shared Volumes are not supported in that release.  However, it also makes note of the following fix. Issue Number: NW112141 Severity: 3 - Low Status: Fixed Problem Summary: Detect if Cluster Shared Volumes (CSVs) are configured on Windows 2008 R2 Additional Information Product: NetWorker Product Feature: Configuration Symptom: Detect if Cluster Shared Volumes (CSVs) are configured on Windows 2008 R2 Workaround or Fix Summary: Fixed in Code Found in Version: 7.5.1 Fixed in Version: 7.5.2 Host OS Vendor: Microsoft Host OS: Windows I cannot find any reference to Cluster Shared Volumes in the documentation for 7.6 at all.  It says nothing about supported or not supported. I have two questions: 1) Are Cluster Shared Volumes (CSV) supported in 7.6? 2) What does "not supported" mean?  That it will not backup the CSV but the rest of the client will backup, that it wil not backup the client successfully, or that it will not install if CSVs are enabled?
My backup server (bserver) is running Win2008x64, Networker 7.5.1 and NMC 7.5.1. My backup client (sqlsrv1) is running Win2003R2-SP2x86, Networker 7.5.1 and NMSQL 5.2.1. If I run a scheduled b... See more...
My backup server (bserver) is running Win2008x64, Networker 7.5.1 and NMC 7.5.1. My backup client (sqlsrv1) is running Win2003R2-SP2x86, Networker 7.5.1 and NMSQL 5.2.1. If I run a scheduled backup of the client's file system, I get readable output in the savegrp.log file on the backup server. Example: ###savegrp.log### Nov 04 19:08:04 bserver: sqlsrv1: C:\ level=full, 9037 MB 01:07:21 45378 files ############### If I run a manual backup of the SQL databases from the client, I can render both the networkr.raw and the nsrsqlsv.raw using nsr_render_log and get readable output on the client. Example: ###Networkr.raw(rendered)### 6211 11/5/2009 10:33:44 AM 2 0 0 6048 5968 0 sqlsrv1.domain.com nwmssql 4690:(pid 4820):BACKUP database [Core] TO virtual_device='Legato#a70553af-9850-2238-85c9-81ac5582bd46' WITH name=N'LegatoNWMSQL' 6211 11/5/2009 10:33:49 AM 2 0 0 6048 5968 0 sqlsrv1.domain.com nwmssql 53085:(pid 4820):Backing up of Core succeeded. ######################### ###nsrsqlsv.raw(rendered)### 37994 11/5/2009 10:33:43 AM 1 0 0 3744 4820 0 sqlsrv1.domain.com (pid4820) Backing up Core... 4690 11/5/2009 10:33:43 AM 1 0 0 3744 4820 0 sqlsrv1.domain.com (pid4820) BACKUP database [Core] TO virtual_device='Legato#a70553af-9850-2238-85c9-81ac5582bd46' WITH name=N'LegatoNWMSQL' 53085 11/5/2009 10:33:49 AM 1 0 0 3744 4820 0 sqlsrv1.domain.com (pid4820) Backing up of Core succeeded. 37996 11/5/2009 10:33:49 AM 1 0 0 3744 4820 0 sqlsrv1.domain.com (pid4820) The backup of Core has completed successfully ######################### However, if I run a scheduled backup of the SQL databases, the lines in the savegrp.log and messages file on the backup server are unreadable. As are the lines in the savegroup completion details in the NMC GUI. They all contain unparsed variables. Example: ###savegrp.log### Nov 05 10:42:47 bserver: * sqlsrv1:MSSQL: 0 1257435749 2 0 0 5248 5388 0 sqlsrv1.domain.com (pid5388) 4 %.*s 2 1 2 50 0 49 NSR_SERVER: bserver.domain.com; Nov 05 10:42:47 bserver: * sqlsrv1:MSSQL: 53085 1257435760 1 0 0 5248 5388 0 sqlsrv1.domain.com (pid5388) 27 Backing up of %s succeeded. 1 0 4 Core Nov 05 10:42:47 bserver: * sqlsrv1:MSSQL: 0 1257435769 1 0 0 5248 5388 0 sqlsrv1.domain.com (pid5388) 54 %s%s: %-*s%s%s%s %*s %2.2ld:%2.2ld:%2.2ld %6ld file(s) 13 0 0 0 6 sqlsrv1 1 2 23 23 6 MSSQL: 0 7 level= 0 4 incr 0 1 , 1 1 9 0 5 62 MB 2 1 0 2 1 0 2 2 20 2 1 4 ################# The savegrp.log and the messages file do not render with nsr_render_log.exe and the *.raw files do not contain any save group progress information. So it looks like the NW base client knows how to parse the variables before sending them to the server but the NMSQL module does not. I have read about this when using older software but all of my installs are the latest release. I remember having this issue in an older NMEXCH module when I updated but then they fixed it with a hotfix. Any ideas on how to get this fixed in the latest release of NMSQL so I can read the log and completion notice on the backup server?
Another question... Can I have two entries in the client's /res/servers file? backupserver1.domain.com backupserver2.domain.com If so, can I just point the client to whatever one I nee... See more...
Another question... Can I have two entries in the client's /res/servers file? backupserver1.domain.com backupserver2.domain.com If so, can I just point the client to whatever one I need for that particular restore? In the beginning, BS1 will have most of the savesets and BS2 will only have the newer ones that it created. BS2 will have more and more as it continues to backup the client. BS1 will have less and less as its retention periods for that client expire and it is no longer taking backups of the client. Thanks, Message was edited by: pdumigan
Thanks Anuj, I wanted to upgrade my existing 7.3.4 to 7.5 and then migrate to new hardware but I was told repeatedly by my EMC sales rep that upgrading from 7.3.4 to 7.5 was not possible and I ... See more...
Thanks Anuj, I wanted to upgrade my existing 7.3.4 to 7.5 and then migrate to new hardware but I was told repeatedly by my EMC sales rep that upgrading from 7.3.4 to 7.5 was not possible and I would have to build my new environment from scratch. Are you now telling me that it would have been possible? My new environment is fully installed, configured and stable now. Tearing it down now would be a considerable waste of the time it took to get it stable. Including support calls, hotfixes and application conflicts (see my other posts). I'm not sure what you mean by "If complete res folder was there". Do you mean if it was copied to the new server? Thanks.
Hi, I have an existing Networker 7.3.4(Win2003) installation where I backup my clients to a backup server. First I backup to disk and then clone full backups to tape monthly that are archived f... See more...
Hi, I have an existing Networker 7.3.4(Win2003) installation where I backup my clients to a backup server. First I backup to disk and then clone full backups to tape monthly that are archived for DR. I have installed and configured an entirely new 7.5 installation. Everything is new including the hardware, the hostname, Win2008, Networker 7.5, licenses, management console, etc... It has been successfully backing itself up and cloning for a while now. Both systems are running beside each other but are unaware of each other's existence. I would like to start migrating clients over to the new system. I would like to move them a couple at a time beginning with the least important to make sure everything is stable. I have read the DR Guide and the suggested method for migration seems to assume that 1) you are migrating tapes and 2) that you have a compelling need to retire the original system. They say to mirror the client names and IDs on the new server, create identical media pools on the new server, use the SCANNER program to inventory the tapes(in my case drives), and recover the client indexes from the scanned tapes/drives. My problems are: 1) making the original system's diskfile available to the new system may be difficult. It will most likely need to be copied over and mounted but it is rather large. If I want to move my clients in sets, I will need to find a way to copy just the deltas as the original system will continue to change. Then I will need to remount and rescan each time I migrate a client. 2) I have no compelling reason to retire the original system immediately. I'm only moving to take advantage of newer software and hardware and because of End of Life schedules for EMC. My questions are: Do I have to migrate the savesets and media? What if I don't do it? Can I just start backing up the client to the new server and stop on the old one? When the save sets on the old server for that client have passed their retention periods, I would remove the client from that server completely. If I start backing up to the new server, will I be able to point the client back to the old server if I need to recover files that were backed up there and are still "live" on the old server. Can a client maintain a good relationship with two servers? Should I use the same client ID on the new server to support archived tapes? Will I be able to scan those tapes in later for DR? Will having the same name and clientID on two servers confuse the client since I may have to point to either server during the transition to recover files from disk? Thanks.
Follow-up to the follow-up... I had to install a networker rollup to troubleshoot another issue. That set the service back to automatic. I had to change it back to delayed start after installin... See more...
Follow-up to the follow-up... I had to install a networker rollup to troubleshoot another issue. That set the service back to automatic. I had to change it back to delayed start after installing the patch.
I tested with 10GB, still the same entries. 0 KB and then finished. Nothing in between. I also tried to clone the same set to the same tape and it just prompted for another tape. Still no change i... See more...
I tested with 10GB, still the same entries. 0 KB and then finished. Nothing in between. I also tried to clone the same set to the same tape and it just prompted for another tape. Still no change in the entries. I agree that constant updates could fill the logs but why report 0 if you are not going to report anything else? (rhetorical) Maybe a percentage update would be better.