@yoshideoeuf undionly.kpxe is only for legacy clients, you must use one from *.efi collection on /tftpboot directory of the fogserver that is working as boot server, I dont know how to do it but maybe you mas configure the windows dhcp server to send different files, on arch dependant
Maybe this links can help you…
@yoshideoeuf can you paste the error?, do you use a proxydhcp or a dhcpd server? have you configure de proxydhcp/dhcp server to provide de efi client? and are you using the correct client?
@sebastian-roth Sugestion tested on production and don’t works, same fail results
@sebastian-roth shrinked version compiling removing unnecesary elements on my instalation. Tested on production and dont works. Boot stuck conecting after boot.php or downloading bzImage/refind*/init.xz. I’ll try your suggestion…
@sebastian-roth I’m not an expert on development, but i’m a multidisciplinary sysadmin, with php knowledges, currently I had made UI changes, system changes, and had on production my own not ended plugin (multihomed location, snapins from webserver, multicast not implemented) (really is based on yours location code), on my little bit big environment, at this moment i think a have time to donate where you think that is needed, I’ll be pleased to make a try, and if it’s convenient and possible make an enrollment… Thanks so much for your great work
@sebastian-roth @george1421 @Fernando-Gietz
Ok at this moment I have a temporal solution compiling on rom-o-matic an shrinked version of the snp.efi and snponly.efi, need to do more tests but at this moment this seems to work…
if it’s possible Iike to see what’s the problem really…
Here are the urls to configure and compile the files…
iPXE file for ACER Travelmate P2
Use this URL to directly retreive your binary for later use:
Editable Configuration URL
Use this URL to adjust your binary’s setup:
iPXE file for dynabook Satellite pro
Use this URL to directly retreive your binary for later use:
Editable Configuration URL
Use this URL to adjust your binary’s setup:
@sebastian-roth
with out debug enable:
error image
https://drive.google.com/file/d/1qimuEy1fm0FuWarHYIpMVu_hlVWEegLR/view?usp=sharing
pcapfile:
https://drive.google.com/file/d/1jp3N9LhPmeZQqQR-rj-Tx2ZfPJW8s3BW/view?usp=sharing
with debug enable too big:
Video of boot proc…
https://drive.google.com/file/d/1BspHvH-nsBWBbhx4uDfofQcSQLj96rWJ/view?usp=sharing
tcpdump result:
https://drive.google.com/file/d/1c7KyZHbCJTbWcfpKNtXak0qNxBSloxyX/view?usp=sharing
With debug always end but it last at more than 15mins downloading files.
@sebastian-roth another notebook with snp.efi, this moel is a Toshiba satellite C40-G-126 same problem stuck downloading bzImage…
on the two models same problem with snponly.efi
with ipxe.efi on this models i retrieved “no configuration methods”, same with realtek.efi
intel.efi not load
I do a bad experiment unloading protocols, and i get into load but with a lot of diferent errors and with a delay of minutes…
Can you suggestme a DEBUG= line to be more clear?
@george1421 @Sebastian-Roth @Fernando-Gietz
On this computer only can start the boot secuence with snp.efi file, but it stucks downloading files like, “refind, refind.conf, bzimage”, I recompile the file using this command and using fog ipxe configuration files and fog ixpe script as you describe on blog:
make bin-x86_64-efi/snp.efi EMBED=/root/git/ipxe-efi/src/ipxescript DEBUG=dhcp,device,efi_driver,efi_init, efi_pci, efi_snp,snp,snponly,snpnet,netdevice,http,tftp,nii
and i get a lot of lines with this error.
NETDEV net0 transmission 0x0 failed: Error 0x2a086089 (http://ipxe.org/2a086089)
https://ipxe.org/err/2a0860
on ipxe.org it tells “Error: No buffer space available”
Can someone to help to fix it?
@mightmar have you verify that the partition scheme is UEFI?
I had that problem, and I verify that this problem can be because the Computer is CSM UEFI+LEGACY, with UEFI first and partition scheme LEGACY (MBR without efi partition), try to configure LAN BOOT OpROM as legacy…
@rodluz you can try this, https://forums.fogproject.org/topic/6463/expose-fog-host-and-image-properties-to-post-install-scripts?_=1620751811503 and make a review of host, image and hostinventory items…
@tesparza if my memory works well we had some little schools working on that way, with all the netboot jailed on local Network, except node data base queries from node to máster/data base, with a few of lesser changes on boot scripts. Foro fogclient need routing to máster…
@george1421 said in InnoDB?:
@Wayne-Workman said in InnoDB?:
So maybe set the default to InnoDB, and we think further on how existing installs can get onto InnoDB?
Now that the OP of the other thread has all of the tables converted I wanted the OP to run for a few weeks under this default configuration (using innodb) and then run the mysqltuner script to see what suggestions it recommends. The current virus situation has delayed this testing both from the FOG Admin side as well as having a full fleet of client computers hitting the fog server. I think changing the database engine to innodb is the first step, the second step is to see what mysqltuner suggests for innodb cache sizes and then make an educated guess where to start from with mysql tuning.
I really feel the sql server and its configuration really needs some attention. It has worked well over the years, so no one really looked at it with a large client base so we could fine tune it. Whenever someone would report high mysql utilization, we would as for them to help debug and they would ghost us after that. So debugging really happened until now.
Hi, invoking an raise dead…
i have my fog database full migrated with out problems, current configuration on this post
https://forums.fogproject.org/topic/14140/bottleneck-on-database/2?_=1602767986457
i have pending optimize the database…
@george1421 I help with this in order i can spend time, currently I’ve too much work but I will take a look at the forum when I can.
@george1421
Excuse the delay, too much work with sars-covid
this is my my.cnf config file.
Current checking time 900s without server stress
When i have a normal situation I make mysqltunner tests to put results/recomendations on this thread
[mysqld]
datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
#wait_timeout = 1
performance_schema = On
skip_name_resolve = On
max_connections = 1000
query_cache_size = 0
query_cache_type = 0
open_files_limit = 44801
table_open_cache = 50000
innodb_log_file_size = 16M
Basic, by the moment with all my workplaces closed i cant optimize database.
I’ve the database on separate virtual server
Ok, What do you want, start and stop VMs from a custom software?
I think that VBox dont support on/off virtual machines in other way than VBox API, Manually or via console tools… But i found on google this post
https://forums.virtualbox.org/viewtopic.php?f=6&t=59776
Tell if it work plz
I do that on physical machines with wol capabilities (turn on) and snapins (turn off).
To wake:
I use 3 options:
1 snapin that do nothing on the script, and mark wake on lan when you launch it.
2 WOL Task.
3 Hardware Inventory WOL checked.
To shutdown system:
I launch an snapin that do nothing with shutdown checked.
I don’t remember if VBox VM’s support WOL, i think no.
As alternative on my virtual servers hosts i have enabled ssh to connect an manualy turn on/off the virtual systems.
Alternatively you can install FOGService on your hosts (are linux based?) on the host and create snapins to turn on/off the virtual machines (Not recommended).
Alternatively if the host are windows it’s possible to do it with powershell remoting, i have it implemented on my nodes to execute some powershell scripts over the network.
Alternatively, you can take a look at the hot picks Hyperbox and phpVirtualBox links on
[https://www.virtualbox.org/](link url)
@EduardoTSeoane
Last report about this…
Everything seems ok, System runing without stops with the same efficiency.
Everything seems to work ok.
Thanks to all
@Sebastian-Roth @fry_p @george1421
Update results:
Currently the server is 8 days uptime.
workload average ±9
workload average max: ±14
checking time = 900s
clients = 17000
Storage nodes = 110
Plugins enabled = Location