FreeBSD 8.2 gets slower over time.

Yes, the real memory is detected each time.
[CMD="grep -i "real memory" /var/log/dmesg.*"][/CMD]
Code: 
/var/log/dmesg.today:real memory  = 6442450944 (6144 MB)
/var/log/dmesg.yesterday:real memory  = 6442450944 (6144 MB)
/var/log/dmesg.yesterday:real memory  = 6442450944 (6144 MB)
/var/log/dmesg.yesterday:real memory  = 6442450944 (6144 MB)

dmidecode:
Code: 
# dmidecode 2.10
SMBIOS 2.4 present.

Handle 0x001B, DMI type 16, 15 bytes
Physical Memory Array
        Location: System Board Or Motherboard
        Use: System Memory
        Error Correction Type: None
        Maximum Capacity: 4 GB
        Error Information Handle: Not Provided
        Number Of Devices: 4

Handle 0x001C, DMI type 17, 27 bytes
Memory Device
        Array Handle: 0x001B
        Error Information Handle: Not Provided
        Total Width: 64 bits
        Data Width: 64 bits
        Size: 1024 MB
        Form Factor: DIMM
        Set: None
        Locator: A0
        Bank Locator: Bank0/1
        Type: Unknown
        Type Detail: None
        Speed: 667 MHz
        Manufacturer:  
        Serial Number:  
        Asset Tag:  
        Part Number:  

Handle 0x001D, DMI type 17, 27 bytes
Memory Device
        Array Handle: 0x001B
        Error Information Handle: Not Provided
        Total Width: 64 bits
        Data Width: 64 bits
        Size: 2048 MB
        Form Factor: DIMM
        Set: None
        Locator: A1
        Bank Locator: Bank2/3
        Type: Unknown
        Type Detail: None
        Speed: 800 MHz
        Manufacturer:  
        Serial Number:  
        Asset Tag:  
        Part Number:  

Handle 0x001E, DMI type 17, 27 bytes
Memory Device
        Array Handle: 0x001B
        Error Information Handle: Not Provided
        Total Width: 64 bits
        Data Width: 64 bits
        Size: 1024 MB
        Form Factor: DIMM
        Set: None
        Locator: A2
        Bank Locator: Bank4/5
        Type: Unknown
        Type Detail: None
        Speed: 667 MHz
        Manufacturer:  
        Serial Number:  
        Asset Tag:  
        Part Number:  

Handle 0x001F, DMI type 17, 27 bytes
Memory Device
        Array Handle: 0x001B
        Error Information Handle: Not Provided
        Total Width: 64 bits
        Data Width: 64 bits
        Size: 2048 MB
        Form Factor: DIMM
        Set: None
        Locator: A3
        Bank Locator: Bank6/7
        Type: Unknown
        Type Detail: None
        Speed: 800 MHz
        Manufacturer:  
        Serial Number:  
        Asset Tag:  
        Part Number:

The system always starts with 6GB memory available in top after a reboot, but over time it disappears. Perhaps I should just try to upgrade to STABLE.
Bios is the latest version.
 
I see you are using modules with different speeds: 667MHz and 800 MHz. This can be a problem - depending on your BIOS settings (what speed did you configure for all modules?).

It is also interesting that BIOS reports maximum size to be 4GB (4x 1GB I guess).
What is the make/model of your motherboard? You can use dmidecode for that too:

# dmidecode --type=2


Strange that 6GB is reported upon boot (so one would suggest this amount is supported).
Did you see any memory related issues in syslog/dmesg?
 
They're already in FreeBSD 8.2-STABLE, if I'm not mistaken?
Anyway the server is still going strong, everything is now fine again :)
 
I don't understand the logic behind deliberately not fixing known bugs in a release, and then putting them in the code after. Basically they released it with known bugs :( standards dropping again :(

If it meant delaying 8.2 for another couple of weeks then do it. I now find myself probably following STABLE again instead of RELEASE as I suspect these wont be patched in via ERRATA.
 
chrcol said:
I dont understand the logic behind deliberatly not fixing known bugs in a release, and then putting them in the code after. Basically they released it with known bugs :( standards dropping again :(
Click to expand...

The whole point of a release is that it's a known state. It's been tested as a unit, everything is synced, it's a fixed point in time. The standards haven't dropped, it's the same as it was. (And "again" is unfair to FreeBSD.)

Of course there will be bugs discovered shortly after release. Or minor bugs might be discovered shortly before release. There isn't any software that is bug-free.

If you want ongoing bug fixes, follow -STABLE. There are a lot of people using it and problems are found and fixed quickly.
 
chrcol said:
I dont understand the logic behind deliberatly not fixing known bugs in a release, and then putting them in the code after. Basically they released it with known bugs :( standards dropping again :(
Click to expand...
Actually it's you that doesn't even have a standard for accuracy. Not fixing bugs found after the tree enters it's slush phase has been SOP for this OS, and pretty much every other as well.

Of course, if you're so dissatisfied with it, you could help testing out earlier and provide some patches.
 
So why haven't patches that have been provided been submitted and made into an ERRATA for the release? Same question really for release 8.1 that was shipped with a zfs bug that stopped booting of the 2nd hdd in mirror when first failed. These are nasty bugs not been fixed in release code. Years ago they the sort of things that used to get patched in.
 
Things don't happen magically, with FreeBSD or any other software. So when "Somebody ought to fix this!" is said, I say "How about you? You're somebody!" So if there are problems, like missing errata, submit a PR. It's one of the great opportunities of FreeBSD that fixing a problem you're having also helps lots of other people.
 
Wblock is right; RELEASE and SNAPSHOT are for testing and building.
If you want a stable system from such, then the mailing lists are where to look.
Volunteering, if you are able, is a good suggestion from wblock.
 
You must log in or register to reply here.
Back
Top