[SOLVED]Re: Bacula Backup Performance
Dave Tibbetts <[email protected]>
| Newsgroups | gmane.comp.bacula.user |
|---|---|
| Message-ID | <[email protected]> |
Thanks everyone for giving me a second set of eyes. I don't know how I completely overlooked my iperf results. When I ran an iperf test from my client to the server, the results showed no retries. But when I ran from the server to the client I had hundreds of retries. I created a new patch cable and ran it to the server. problem solved. Backups back to normal. Thanks again! Dave On 9/18/25 10:48 AM, Dave Tibbetts wrote: > Rob, > Thanks man. yeah, I totally spaced on this. I'm going to check the > cable, but I've got a bunch of Cat-5E and I might just create a new > cable and go from there. > > Thank you everyone. I owe you all a coffee! > > Dave > On 9/18/25 10:25 AM, Rob Gerber wrote: >> Whoops. I missed that. Apparently Dragan is the only one of us who >> has had their coffee today. >> >> Dragan is 100% correct. Your network speeds indicate that you have a >> fast ethernet (10/100, max 100 megabits per second) network link >> somewhere. As opposed to a gigabit ethernet (10/100/1000, max 1000 >> megabits per second, aka, 1 gbps) network. >> >> Your speeds, as determined by iperf, clearly indicate that your >> network isn't capable of going any faster than fast ethernet / >> 100mbps speeds, at least between your desktop and your bacula server. >> Those aren't solid speeds. They're rookie numbers, unless you're >> stuck on really old hardware. >> >> Fun fact: fast ethernet was designed to capable of working with only >> wires 1,2,3,6 out of 1-8 in a category 5-6 ethernet cable. >> >> Further fun fact: gigabit ethernet requires all 8 wires to function >> properly. >> >> Final fun fact: gigabit network cards are usually capable of falling >> back to 100 mbps rates if necessary. Like if, say, one of the wires >> 4,5,7,8 were disconnected in some cable. >> >> I think you might have a bad patch cable. Probably between the server >> and everything else. >> >> Either way, start checking switches, nice ports, etc for any ports >> that are in fast ethernet mode. Some switches change the color of the >> link lights from green to orange, if a port is operating in FE mode. >> >> Given the sudden nature of it, I really suspect a patch cable. >> >> If you have structured cabling, a Klein tools VDV Scout Pro network >> tester is dead simple to use, and reasonably affordable. >> >> Robert Gerber >> 402-237-8692 >> [email protected] >> >> On Thu, Sep 18, 2025, 8:55 AM Dragan Milivojević >> <[email protected]> wrote: >> >> If your network speed is 95 M BITS and bacula transfer rate is 11 >> M BYTES ... >> >> On Thu, 18 Sept 2025 at 15:24, Dave Tibbetts <[email protected]> >> wrote: >> > >> > Another slow backup issue and I'm hoping someone can help. >> > >> > I'm running Bacula 13.04 on Ubuntu 25.04. >> > It's been running fine for several years over many Bacula and >> OS upgrades. >> > >> > Suddenly on 3 August 2025, my backup speed was cut by about >> half. I have a desktop that typically would run about 25 to 26 >> MB/s and is now running at 11.5. This is running via ethernet. >> > >> > I'm seeing the same with another machine that's also on >> ethernet and a wireless laptop. >> > >> > I'm using virtual tapes in the filestor, and the bacula >> director and storage are on the same machine. Again, small setup >> and been very reliable. >> > >> > SpoolAttributes is set on for all jobs. >> > >> > I ran iperf from my desktop to my server and got solid numbers: >> > [ ID] Interval Transfer Bitrate Retr Cwnd >> > [ 5] 0.00-1.00 sec 12.4 MBytes 104 Mbits/sec 0 255 >> KBytes >> > [ 5] 1.00-2.00 sec 11.4 MBytes 95.5 Mbits/sec 0 255 >> KBytes >> > [ 5] 2.00-3.00 sec 11.0 MBytes 92.3 Mbits/sec 0 255 >> KBytes >> > [ 5] 3.00-4.00 sec 11.5 MBytes 96.4 Mbits/sec 0 255 >> KBytes >> > [ 5] 4.00-5.00 sec 10.9 MBytes 91.2 Mbits/sec 0 255 >> KBytes >> > [ 5] 5.00-6.00 sec 11.5 MBytes 96.5 Mbits/sec 0 255 >> KBytes >> > [ 5] 6.00-7.00 sec 10.9 MBytes 91.2 Mbits/sec 0 255 >> KBytes >> > [ 5] 7.00-8.00 sec 11.5 MBytes 96.5 Mbits/sec 0 255 >> KBytes >> > [ 5] 8.00-9.00 sec 11.4 MBytes 95.4 Mbits/sec 0 255 >> KBytes >> > [ 5] 9.00-10.00 sec 11.0 MBytes 92.2 Mbits/sec 0 255 >> KBytes >> > - - - - - - - - - - - - - - - - - - - - - - - - - >> > [ ID] Interval Transfer Bitrate Retr >> > [ 5] 0.00-10.00 sec 113 MBytes 95.1 Mbits/sec 0 >> sender >> > [ 5] 0.00-10.02 sec 112 MBytes 94.1 Mbits/sec >> receiver >> > >> > I tried using ethtool to tweak the NIC's - >> > >> > ethtool -K eno1 tso off gso off gro off >> > >> > But that didn't help. >> > >> > >> > I went back through the Bacula archives and tried many, many >> different things - none of which helped. I finally decided to >> just delete the rebuild the Postgresql database and delete all >> the virtual tapes. >> > >> > So I brought everything back up and started a backup on a brand >> new catalog. Same throughput 11.5 MB/s >> > Bugs-Bunny-dir JobId 1: End auto prune. >> > Bugs-Bunny-dir JobId 1: No Files found to prune. >> > Bugs-Bunny-dir JobId 1: Begin pruning Files. >> > Bugs-Bunny-dir JobId 1: No Jobs found to prune. >> > Bugs-Bunny-dir JobId 1: Begin pruning Jobs older than 1 month >> 16 days . >> > Bugs-Bunny-dir JobId 1: Bacula Enterprise Bugs-Bunny-dir 13.0.4 >> (12Feb24): >> > Build OS: x86_64-pc-linux-gnu-bacula-enterprise ubuntu 22.04 >> > JobId: 1 >> > Job: Backup_Bullwinkle.2025-09-17_15.58.07_17 >> > Backup Level: Full (upgraded from Incremental) >> > Client: "Bullwinkle-fd" 15.0.3 (25Mar25) >> x86_64-redhat-linux-gnu,redhat,; >> > FileSet: "Bullwinkle Set" 2025-09-17 15:58:07 >> > Pool: "File" (From Command input) >> > Catalog: "MyCatalog" (From Client resource) >> > Storage: "Filestor" (From Command input) >> > Scheduled time: 17-Sep-2025 15:58:07 >> > Start time: 17-Sep-2025 15:58:09 >> > End time: 18-Sep-2025 06:48:23 >> > Elapsed time: 14 hours 50 mins 14 secs >> > Priority: 10 >> > FD Files Written: 145,410 >> > SD Files Written: 145,410 >> > FD Bytes Written: 618,967,362,190 (618.9 GB) >> > SD Bytes Written: 618,992,689,163 (618.9 GB) >> > Rate: 11588.1 KB/s >> > Software Compression: 2.5% 1.0:1 >> > Comm Line Compression: 5.2% 1.1:1 >> > Snapshot/VSS: no >> > Encryption: no >> > Accurate: no >> > Volume name(s): >> Oviedo0001|Oviedo0002|Oviedo0003|Oviedo0004|Oviedo0005|Oviedo0006|Oviedo0007|Oviedo0008|Oviedo0009|Oviedo0010|Oviedo0011|Oviedo0012|Oviedo0013|Oviedo0014|Oviedo0015|Oviedo0016|Oviedo0017|Oviedo0018|Oviedo0019|Oviedo0020|Oviedo0021|Oviedo0022|Oviedo0023|Oviedo0024|Oviedo0025|Oviedo0026|Oviedo0027|Oviedo0028|Oviedo0029|Oviedo0030|Oviedo0031|Oviedo0032|Oviedo0033|Oviedo0034|Oviedo0035|Oviedo0036|Oviedo0037|Oviedo0038|Oviedo0039| >> > Volume Session Id: 1 >> > Volume Session Time: 1758138994 >> > Last Volume Bytes: 7,428,533,207 (7.428 GB) >> > Non-fatal FD errors: 0 >> > SD Errors: 0 >> > FD termination status: OK >> > SD termination status: OK >> > Termination: Backup OK >> > Bugs-Bunny-sd JobId 1: Sending spooled attrs to the Director. >> Despooling 40,737,115 bytes ... >> > Bugs-Bunny-sd JobId 1: Elapsed time=14:49:57, Transfer >> rate=11.59 M Bytes/second >> > >> > Is it something crazy like my FD on the desktop is 15.0.3 and >> the server is running 13.04? >> > >> > I'm just totally out of ideas. What am I missing? >> > >> > Thanks, >> > Dave >> > >> > >> > >> > _______________________________________________ >> > Bacula-users mailing list >> > [email protected] >> > https://lists.sourceforge.net/lists/listinfo/bacula-users >> >> >> _______________________________________________ >> Bacula-users mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/bacula-users >> > _______________________________________________ Bacula-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/bacula-users