FastTrack

Starting from v6.29rc9, we introduced new and existing feature – FastTrack. Easy way to make your Firewall/NAT router up to 5x faster.

*) ipv4 fasttrack fastpath – accelerates connection tracking and nat for marked connections (more than 5x performance improvement compared to regular slow path conntrack/nat) – currently limited to TCP/UDP only;
*) added ~fasttrack-connection~ firewall action in filter/mangle tables for marking connections as fasttrack;
*) added fastpath support for bridge interfaces – packets received and transmitted on bridge interface can go fastpath (previously only bridge forwarded packets could go fastpath);
*) packets now can go half-fastpath – if input interface supports fastpath and packet gets forwarded in fastpath but output interface does not support fastpath or has interface queue other than only-hw-queue packet gets converted to slow path only at the dst interface transmit time;
*) trafflow: add natted addrs/ports to ipv4 flow info.

http://www.mikrotik.com/download/share/FastTrack.pdf

 

http://forum.mikrotik.com/viewtopic.php?t=96302

 

 

 

 

分類: RouterOS | 發佈留言

qcow2–>raw

VMDK–>qcow2:

kevin@kevin:~# qemu-img convert -f vmdk -O qcow2 SLES11SP1-single.vmdk SLES11SP1-single.img

http://www.ibm.com/developerworks/cn/linux/l-cn-mgrtvm3/index.html

qcow2–>raw:

kevin@kevin:~$ qemu-img convert -O qcow2 image-raw.raw image-raw-converted.qcow

本文转自:http://blog.prajnagarden.com/?p=248

閱讀全文 →

分類: RouterOS | 發佈留言

To create a qemu guest with more than one nic, just repeat the -net commands

To create a qemu guest with more than one nic, just repeat the -net commands

qemu -net tap,vlan=0,ifname=tap0,script=./qemu-ifup -net nic,vlan0 \ -net tap,vlan=0,ifname=tap1,script=./qemu-ifup -net nic,vlan0 \ -net tap,vlan=0,ifname=tap2,script=./qemu-ifup -net nic,vlan0 \ -boot d -cdrom alpine*.iso}}

 

 

 

qemu-system-x86_64 rhel6u3.img -smp 2 -m 1024 -net nic -nettap,ifname=tap1,script=/etc/qemu-ifup,downscript=no -vnc :0 -daemonize


							
分類: RouterOS | 發佈留言

從無到有,搞定你的RouterOS讓它能連上網路.

假如你的RouterOS是x86安裝的,或是可能某些硬體沒有出廠的預設值,

底下的圖文教學協助你讓你的RouterOS可以連上網路並透過它上網.

如果是routerboard系列的原廠機器,現在不需要這麼麻煩了,

原廠內建快速設定頁面,新機器只需要透過瀏覽器或是WINBOX填入相關的資訊即可連線上網.

閱讀全文 →

分類: RouterOS | 7 則留言

UniFi Controller AC控制器安裝SSL證書

現在有另外一種更簡單的方法了~

參考這裡網址

 

 

———————–以下是舊文————————

解决每次開啟UniFi Controller AC控制器都會出現

您的連線不是私人連線

的安全性警告畫面.

如底下畫面.

閱讀全文 →

分類: Ubnt 無線 | 1 則留言

Proxmox 管理虛擬機器 使用「qm」指令 

Proxmox 管理虛擬機器 使用「qm」指令

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
qm <command> <vmid> [OPTIONS]
qm [create|set] <vmid>
    --memory  <MBYTES>    memory in MB (64 - 8192)
    --sockets <N>         set number of CPU sockets <N>
    --cores <N>           set cores per socket to <N>
    --ostype NAME         specify OS type
    --onboot [yes|no]     start at boot
    --keyboard XX         set vnc keyboard layout
    --cpuunits <num>      CPU weight for a VM
    --name <text>         set a name for the VM
    --description <text>  set VM description
    --boot [a|c|d|n]      specify boot order
    --bootdisk <disk>     enable booting from <disk>
    --acpi (yes|no)       enable/disable ACPI
    --kvm (yes|no)        enable/disable KVM
    --tdf (yes|no)        enable/disable time drift fix
    --localtime (yes|no)  set the RTC to local time
    --vga (gd5446|vesa)   specify VGA type
    --vlan[0-9u]          MODEL=XX:XX:XX:XX:XX:XX[,MODEL=YY:YY:YY:YY:YY:YY]
    --ide<N>              [volume=]volume,[,media=cdrom|disk]
                          [,cyls=c,heads=h,secs=s[,trans=t]]
                          [,cache=none|writethrough|writeback]
                          [,snapshot=on|off][,cache=on|off][,format=f]
                          [,werror=enospc|ignore|report|stop]
                          [,rerror=ignore|report|stop]
                          [,backup=no|yes]
    --ide<N> <GBYTES>     create new disk
    --ide<N> delete       remove drive - destroy image
    --ide<N> undef        remove drive - keep image
    --cdrom <file>        is an alias for --ide2 <file>,media=cdrom
    --scsi<N>             [volume=]volume,[,media=cdrom|disk]
                          [,cyls=c,heads=h,secs=s[,trans=t]]
                          [,snapshot=on|off][,format=f]
                          [,cache=none|writethrough|writeback]
                          [,werror=enospc|ignore|report|stop]
                          [,backup=no|yes]
    --scsi<N> <GBYTES>    create new disk
    --scsi<N> delete      remove drive - destroy image
    --scsi<N> undef       remove drive - keep image
    --virtio<N>           [volume=]volume,[,media=cdrom|disk]
                          [,cyls=c,heads=h,secs=s[,trans=t]]
                          [,snapshot=on|off][,format=f]
                          [,cache=none|writethrough|writeback]
                          [,werror=enospc|ignore|report|stop]
                          [,rerror=ignore|report|stop]
                          [,backup=no|yes]
    --virtio<N> <GBYTES>  create new disk
    --virtio<N> delete    remove drive - destroy image
    --virtio<N> undef     remove drive - keep image
qm monitor <vmid>       connect to vm control monitor
qm start <vmid>         start vm
qm shutdown <vmid>      gracefully stop vm (send poweroff)
qm wait <vmid> [time]   wait until vm is stopped
qm stop <vmid>          kill vm (immediate stop)
qm reset <vmid>         reset vm (stop, start)
qm suspend <vmid>       suspend vm
qm resume <vmid>        resume vm
qm cad <vmid>           sendkey ctrl-alt-delete
qm destroy <vmid>       destroy vm (delete all used/owned volumes)
qm unlock <vmid>        clear migrate/backup lock
qm status <vmid>        shows the container status
qm cdrom <vmid> [<device>] <path>  set cdrom path. <device is ide2 by default>
qm cdrom <vmid> [<device>] eject   eject cdrom
qm unlink <vmid> <volume>  delete unused disk images
qm vncproxy <vmid> <ticket>  open vnc proxy
qm vnc <vmid>           start (X11) vncviewer (experimental)
qm showcmd <vmid>       show command line (debug info)
qm list                 list all virtual machines
qm startall             start all virtual machines (when onboot=1)
qm stopall [timeout]    stop all virtual machines (default timeout is 3 minutes)
分類: RouterOS | 發佈留言

Queues – PCQ Examples

Manual:Queues – PCQ Examples

Per Connection Queue (PCQ) is a queuing discipline that can be used to dynamically equalize or shape traffic for multiple users, using little administration. It is possible to divide PCQ scenarios into three major groups: equal bandwidth for a number of users, certain bandwidth equal distribution between users, unknown bandwidth equal distribution between users.

Equal Bandwidth for a Number of Users

Use PCQ type queue when you need to equalize the bandwidth [and set max limit] for a number of users. We will set the 64kbps download and 32kbps upload limits.

PCQ.png

There are two ways how to make this: using mangle and queue trees, or, using simple queues.

1. Mark all packets with packet-marks upload/download: (lets constider that ether1-LAN is public interface to the Internet and ether2-LAN is local interface where clients are connected

/ip firewall mangle add chain=prerouting action=mark-packet \
   in-interface=ether1-LAN new-packet-mark=client_upload
/ip firewall mangle add chain=prerouting action=mark-packet \
   in-interface=ether2-WAN new-packet-mark=client_download

2. Setup two PCQ queue types – one for download and one for upload. dst-address is classifier for user’s download traffic, src-address for upload traffic:

/queue type add name="PCQ_download" kind=pcq pcq-rate=64000 pcq-classifier=dst-address
/queue type add name="PCQ_upload" kind=pcq pcq-rate=32000 pcq-classifier=src-address

3. Finally, two queue rules are required, one for download and one for upload:

/queue tree add parent=global-in queue=PCQ_download packet-mark=client_download
/queue tree add parent=global-out queue=PCQ_upload packet-mark=client_upload

If you don’t like using mangle and queue trees, you can skip step 1, do step 2, and step 3 would be to create one simple queue as shown here:

/queue simple add target-addresses=192.168.0.0/24 queue=PCQ_upload/PCQ_download
分類: RouterOS | 發佈留言

[Ubuntu] ConohaのVPSのUbuntuで2つ目のHDDをマウントする

[Ubuntu] ConohaのVPSのUbuntuで2つ目のHDDをマウントする

ConohaのVPSでは、ディスク領域がDisk1, Disk2の2つのパーティションに分割され、Disk1にOSがインストールされます。

デフォルトのCent OSを利用する場合は、Disk2が/dataに自動でマウントされるようですが、Ubuntuを上書きでインストールした場合にはDisk2が自動ではマウントされませんでした。

以下のコマンドを実行することで、Disk2をマウント可能です。

1. Disk2のフォーマット

Disk2は通常/dev/vdbとして認識されているので、これをext4形式にフォーマットします。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
$ sudo mkfs.ext4 /dev/vdb
mke2fs 1.42.9 (4-Feb-2014)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
6553600 inodes, 26214400 blocks
1310720 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=4294967296
800 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624, 11239424, 20480000, 23887872
Allocating group tables: done
Writing inode tables: done
Creating journal (32768 blocks): done
Writing superblocks and filesystem accounting information: done

2. マウントできるかを確認

フォーマットしたDisk2を/dataにマウントできるかを確認します。

1
2
3
4
5
6
7
8
9
10
11
12
$ sudo mkdir /data
$ sudo mount /dev/vdb /data/
$ df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/vda1        97G  1.9G   90G   3% /
none            4.0K     0  4.0K   0% /sys/fs/cgroup
udev            991M  4.0K  991M   1% /dev
tmpfs           201M  380K  200M   1% /run
none            5.0M     0  5.0M   0% /run/lock
none           1001M     0 1001M   0% /run/shm
none            100M     0  100M   0% /run/user
/dev/vdb         99G   60M   94G   1% /data

自動マウント設定をするため、一旦アンマウントします。

1
$ sudo umount /data

3. 起動時に自動でマウントされるように設定

Disk2のUUIDを確認します。

1
2
3
4
$ sudo -i blkid
/dev/vda1: UUID="UUIDその1" TYPE="ext4"
/dev/vda5: UUID="UUIDその2" TYPE="swap"
/dev/vdb: UUID="このUUIDをコピー" TYPE="ext4"

fstabに追記します。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
$ sudo vi /etc/fstab
# /etc/fstab: static file system information.
#
# Use 'blkid' to print the universally unique identifier for a
# device; this may be used with UUID= as a more robust way to name devices
# that works even if disks are added and removed. See fstab(5).
#
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
# / was on /dev/vda1 during installation
UUID=UUIDその1                            /               ext4    errors=remount-ro 0       1
# swap was on /dev/vda5 during installation
UUID=UUIDその2                            none            swap    sw              0       0
# データ領域のマウント
UUID=先ほどコピーしたUUID                 /data           ext4    defaults        1       2
分類: RouterOS | 發佈留言

MySql 修改 root 密碼 (忘記密碼)

必須重新啟動 mysql 並且進入安全模式,即可以root不需要 密碼登入。

[root@ros~]# service mysql  stop
[root@ros~]# mysqld_safe –user=mysql –skip-grant-tables –skip-networking &
mysql> UPDATE user SET Password=PASSWORD(‘這裡填你要重設的密碼’) where USER=‘root’;
mysql> FLUSH PRIVILEGES;
mysql> quit
搞定

 

閱讀全文 →
分類: RouterOS | 發佈留言

RB系列Master-Port作用(舊版OS v6.41rc1以前)

(目前新版自v6.41rc1以後已經不用Master-Port的方式,已全部改由Bridge方式設定橋接)

如果你是新版的系統,請直接忽略這篇文章.

閱讀全文 →

分類: RouterOS | 4 則留言