The error clearly says your PC disk is fulltried everything and keep getting this error
The error clearly says your PC disk is fulltried everything and keep getting this error
I'm getting same error here....tried everything and keep getting this error
that didnt work i did try it as you say with the swap filesRight-click the VM and go to Edit Settings > Options tab > Swapfile location. Select one of the other options ("Always store with the virtual machine" is recommended and will probably work for you since that datastore has a lot of free space). As far as cleaning up that datastore, you can right-click it and Browse and poke around to see what kind of files are taking up the space. It may be likely that your host is configured to use a different datastore for all VMs by default and it is just taken up by swap files at this point.
ploquets - there an issue on ESXi and XVM (including Amazon) that when you want to import the installer image that is provided on the download page you have to run through the installer and then that image should be used in AMI or virtual machine creation on XEN, ESXi or Amazon (that is XEN)
DataStore has plenty space... datastore has a lot of free space). As far as cleaning up that datastore, you can right-click it and Browse and poke around to see what kind of files are taking up the space. It may be likely that your host is configured to use a different datastore for all VMs by default and it is just taken up by swap files at this point.
Hello dirtonth,Hey Guys,
I am using CHR on AWS using the AMI you provided, and upgraded it to the latest RC30.
Everytime the router is restarted, the hostname changes to the reverse-dns. Is there a way of blocking this from happening ?
Thanks,
Matthew.
The only way around that issue with ESXi that I have found so far is to first run the disk in VMware Workstation and let it install. Then export the working VM to an OVA or OVF and use that via "deploy ova" in the Vsphere client. It's the only way I have been able to successfully deploy CHR on ESXi.DataStore has plenty space... datastore has a lot of free space). As far as cleaning up that datastore, you can right-click it and Browse and poke around to see what kind of files are taking up the space. It may be likely that your host is configured to use a different datastore for all VMs by default and it is just taken up by swap files at this point.
Even after changing the swap files for every possible option, I get same error.
Seems that the VMware snapshot engine is causing this, because the error is at the .s001 which seems to be linked with the snapshot system.
But I've tried to make the disk Independent (supposedly will disable snapshot engine) but it fails again.
there are problems on Windows 10 Pro Hyper-V, probably you have the same problem on your version of Windows. We are working on the issue for Windows 10 Pro at the moment. From get go we tested on Windows server.Just downloaded chr-6.34rc30.vhdx. It looks like it gets stuck trying to start services and reboots itself.
Windows 8.1 - Hyper-v
VM has 2GB of ram (tried 512MB also)
Anyone have a link to an older RC that I can try ?
*Edit*
Scratch that, looks like RC 30,29,21 all seem to reboot continuously. Anyone have a recommendation?
*Edit 2*
Scratch all of this. It looks like the CHR will not boot / function on my laptop (Win 8.1 64bit) with Hyper-v (possibly something broken on my end). Looks like the VM boots on a test bed server 2012R2 + Hyper-v.
The only way around that issue with ESXi that I have found so far is to first run the disk in VMware Workstation and let it install. Then export the working VM to an OVA or OVF and use that via "deploy ova" in the Vsphere client. It's the only way I have been able to successfully deploy CHR on ESXi.DataStore has plenty space... datastore has a lot of free space). As far as cleaning up that datastore, you can right-click it and Browse and poke around to see what kind of files are taking up the space. It may be likely that your host is configured to use a different datastore for all VMs by default and it is just taken up by swap files at this point.
Even after changing the swap files for every possible option, I get same error.
Seems that the VMware snapshot engine is causing this, because the error is at the .s001 which seems to be linked with the snapshot system.
But I've tried to make the disk Independent (supposedly will disable snapshot engine) but it fails again.
[/quote]The only way around that issue with ESXi that I have found so far is to first run the disk in VMware Workstation and let it install. Then export the working VM to an OVA or OVF and use that via "deploy ova" in the Vsphere client. It's the only way I have been able to successfully deploy CHR on ESXi.DataStore has plenty space... datastore has a lot of free space). As far as cleaning up that datastore, you can right-click it and Browse and poke around to see what kind of files are taking up the space. It may be likely that your host is configured to use a different datastore for all VMs by default and it is just taken up by swap files at this point.
Even after changing the swap files for every possible option, I get same error.
Seems that the VMware snapshot engine is causing this, because the error is at the .s001 which seems to be linked with the snapshot system.
But I've tried to make the disk Independent (supposedly will disable snapshot engine) but it fails again.
I cant make it start with xen 4.4I also want to use CHR with XenServer 6.5.0 but doesn't start also with viridian=false... what we need to do?
According to meWhy call it "Cloud Hosted"? Is there the ability to actually have a custom RouterOS installation hosted on a general hosting provider (NOT a specialized provider that would let you run custom VMs...)?
Have you guys (at MikroTik) considered maybe offering such a service (preferably one which, just like this VM, has full functionality, but with rate limits). If you do, I for one would certainly end up using it, as right now, to test my API client, I need to set up a VM on my PC, and no 3rd party CI service allows me to do the same with their infrastructure.
As it currently stands, this "Cloud Hosted Router" is really more like "RouterOS VM Edition".
what you did? Did you try to start KVM using this image so that installer finishes and then run XEN using this.There is very large number of people asking for XENSERVER support.
Anyone knows how to fix this?
http://puu.sh/mvStA/c2dfa0360e.png
No, I dont know how KVM works.what you did? Did you try to start KVM using this image so that installer finishes and then run XEN using this.There is very large number of people asking for XENSERVER support.
Anyone knows how to fix this?
http://puu.sh/mvStA/c2dfa0360e.png
Its workingNo, I dont know how KVM works.what you did? Did you try to start KVM using this image so that installer finishes and then run XEN using this.There is very large number of people asking for XENSERVER support.
Anyone knows how to fix this?
http://puu.sh/mvStA/c2dfa0360e.png
I tried kvm img but it fails.
That error can be fixed runing kvm on it?
I got a working img in xen.4.4 , should I try import that image?
Hello mpreissner,Anyone have performance metrics related to AWS instance size? I'm looking at using the CHR as a VPN/Router for a corporate cloud infrastructure, and want to be able to price out the instance sizes in terms of routing performance within the VPC as well as number of concurrent users I can support via VPN (most likely SSTP).
Any info is greatly appreciated!
Hello mpreissner,Anyone have performance metrics related to AWS instance size? I'm looking at using the CHR as a VPN/Router for a corporate cloud infrastructure, and want to be able to price out the instance sizes in terms of routing performance within the VPC as well as number of concurrent users I can support via VPN (most likely SSTP).
Any info is greatly appreciated!
I just launched tree t2.nano instances in VPC in production environment. The performance is very nice. The maximum throughput is like 150 Mbps using that kind of instance (smallest one in AWS). I am using as VPN server and gateway, and also running BGP.
I am very happy with the stability and performance.
I asume that the performance will increase if you use another instance like m3, or bigger.
Best regards.
Sent from my SM-A500Y using Tapatalk
Eventually, this will become available for other architectures too. So yeah, 6.34rc39 for CHR fixes important problem on AWS. So get this if you are launching new instances.@Janisk
For now the 6.34rc39 version is not available. The latest is 6.34rc36.
AtmanActive, am I understanding you correctly that even after generating new System ID it remains the same 'WtgfTcn7YNA'?Hello,
... I get the error message 'ERROR: This System ID with old ID already received trial license key' on the Renew License window where I need to type in my MikroTik shop account username and password.
License window shows System ID 'WtgfTcn7YNA', level 'free', Next Renewal At:'', Deadline At:'', but even when I try 'Generate New ID', still I get the same result.
HiI'm trying to use a CHR as PPPoE Server with Radius authentication.
But every time a client reach a full traffic specified on queue... the CHR reboots.
This happens specially when using a SpeedTest.
If I kill the queue related to the PPPoE client, everything works OK. But then I cannot limit the client.
The queue is generated automatically, but even if I try to setup manually the same kind of queue, same thing happens.
Using latest version 6.34rc34
Please Help!
------------------------------------------------------------------------------------------------------------------------------------
Edit:
I've installed the stable version, 6.33.3 x86 on top of ESXi and everything went well.
Same scenario from the CHR, but CHR reboots it self when queue reaches the top.
It's probably a bug from CHR. Thanks
create support output file on running, then shut down/start and if interface name has changed again - create another support output file and send it to us so we can check what is the issue with this.Ok some more simple tests today in azure, (thanks to support for the test license)
Getting good bandwidth on receive tests inside Azure,
634 Mbps
Having some problems on send tests im looking at, ill update soon
One thing worth noting, every shutdown and restart enables a new interface, so after 7 startups I am on ether7 for my lan interface. This is probably udev related, ive seen this on centos on azure before.
https://azure.microsoft.com/en-us/docum ... hd-centos/
Going to try some vpn tunnels next, I have BGP already runnning over an openvpn tunnel between AWS & AZURE (two private networks, AWS ), very stable & low latency.
What is the trick to get it running in XenServer? All of us are stuck with the dreaded "Booting kernel failed: Invalid argument" message.Got it running on our XenServer 6.5 environment (CHR 6.34rc41)
right now I only see the images on the download page, no 'upgrade' packages.syadnom
whan you mean - upgradable? It should be from the start upgradable to newer version. Only when you run out of the lincese time upgrade is not possible.
Current:right now I only see the images on the download page, no 'upgrade' packages.syadnom
whan you mean - upgradable? It should be from the start upgradable to newer version. Only when you run out of the lincese time upgrade is not possible.
I'm on rc41, but rc45 is available. How do I upgrade?
/system package update
set channel=current
check-for-updates
/system package update
set channel=current
download
/system reboot
/system package update
set channel=release-candidate
check-for-updates
/system package update
set channel=release-candidate
download
/system reboot
I downloaded the VDI image, then made a new VM in VirtualBox with that VDI as hard disk (according howto on page 1). Started the machine in VirtualBox to let the image install. Then exported the machine and imported it in XenServer 6.5 with the profile 'Other install media' and voila it boots!What is the trick to get it running in XenServer? All of us are stuck with the dreaded "Booting kernel failed: Invalid argument" message.Got it running on our XenServer 6.5 environment (CHR 6.34rc41)
I just did it once and have imported the same 'bare image' a bunch of times. Then generate a new ID once you've booted it up. Now that I figured out (...was told what to do...) updating, this is really simple and takes just 30-40 seconds to spawn a new CHR from the template.Any updates on making an OVA so it doesn't take 47 steps to get into vSphere?
I could only find the ami-5081a53a in us-east-1, could not find anything in us-west-1 or us-west-2.for amazon cloud - you have to update to 6.34rc39 or newer.
Here is the new AMI IDs (~1GB of storage):
AMI ID us-west1 ami-8fba10fc (on us-east1 ami-5081a53a)
p.s. if there is important update we will create new AMI image so it is easier to launch as generic CHR as it is possible.
HiHiI'm trying to use a CHR as PPPoE Server with Radius authentication.
But every time a client reach a full traffic specified on queue... the CHR reboots.
This happens specially when using a SpeedTest.
If I kill the queue related to the PPPoE client, everything works OK. But then I cannot limit the client.
The queue is generated automatically, but even if I try to setup manually the same kind of queue, same thing happens.
Using latest version 6.34rc34
Please Help!
------------------------------------------------------------------------------------------------------------------------------------
Edit:
I've installed the stable version, 6.33.3 x86 on top of ESXi and everything went well.
Same scenario from the CHR, but CHR reboots it self when queue reaches the top.
It's probably a bug from CHR. Thanks
I use CHR 6.34rc41 on linux kvm ubuntu 14.04
If I enable mangling rules and queue tree, routeros reboots on queue limit. Seems to be the same bug.
Kind regards
try with at least 128MB of RAM.Using CHR 6.34 on Qemu 2.4.0, I can duplicate the error eugenics61 mentions above with the vmdk, vdi and img images.
Huh... It worked.try with at least 128MB of RAM.Using CHR 6.34 on Qemu 2.4.0, I can duplicate the error eugenics61 mentions above with the vmdk, vdi and img images.
do you have support output file generated for this setup, if so, then send it to support.I'm trying to use a CHR as PPPoE Server with Radius authentication.
But every time a client reach a full traffic specified on queue... the CHR reboots.
This happens specially when using a SpeedTest.
If I kill the queue related to the PPPoE client, everything works OK. But then I cannot limit the client.
The queue is generated automatically, but even if I try to setup manually the same kind of queue, same thing happens.
Using latest version 6.34rc34
Please Help!
------------------------------------------------------------------------------------------------------------------------------------
Edit:
I've installed the stable version, 6.33.3 x86 on top of ESXi and everything went well.
Same scenario from the CHR, but CHR reboots it self when queue reaches the top.
It's probably a bug from CHR. Thanks
I get the following when trying to copy the us-east1 AMI:btw, updated AMI image to the 6.34 new AMI IDs are as follows for the regions eu-west1 ami-7d77c10e, on us-east1 ami-07ddf26d
You can copy the AMI to the other region where you reside.
sorry about that. Snapshots are made public now for US N.Arizona and EU Irelan clusters.
I just noticed that no matter what you set the disk size to on AWS EC2 when launching the official AMI(or copied), it seems to be only around 100MB...sorry about that. Snapshots are made public now for US N.Arizona and EU Irelan clusters.
Hyper-v has given me issues setting the hardware time on Linux vms. I suspect this is just hyper-v being peculiar.Just installing my first install CHR and playing with it. Hyper-V host.
The time service is off. It seems to take the time from the host. But it ignores the datetime offset.
If I configure CHR timezone, the offset gets applied twice Quite obvious for me, as I am in europe, the machine in chicago and the server is running US local time.
In order to get the correct time, I must set it to Etc/GMT+0 - which neither me nor the machine have.
Do you try to use the Hyper-V "Integration feature" of "Time synchronization"? try setting up VMs with that on and off and see if there is a difference. I never have time issues with Time Synchronization on, but I only use Centos 6/7 on Hyper-V, and I always install the hyper-v tools in the guest via yum.Hyper-v has given me issues setting the hardware time on Linux vms. I suspect this is just hyper-v being peculiar.Just installing my first install CHR and playing with it. Hyper-V host.
The time service is off. It seems to take the time from the host. But it ignores the datetime offset.
If I configure CHR timezone, the offset gets applied twice Quite obvious for me, as I am in europe, the machine in chicago and the server is running US local time.
In order to get the correct time, I must set it to Etc/GMT+0 - which neither me nor the machine have.
my bad. Here are new AMIs with 6.34.1 and proper image size: eu-west1 ami-bef141cd; us-east1 ami-3f486355I just noticed that no matter what you set the disk size to on AWS EC2 when launching the official AMI(or copied), it seems to be only around 100MB...sorry about that. Snapshots are made public now for US N.Arizona and EU Irelan clusters.
[admin@] > /system resource print
uptime: 3m39s
version: 6.34 (stable)
build-time: Jan/29/2016 10:25:47
free-memory: 479.0MiB
total-memory: 496.3MiB
cpu: Intel(R)
cpu-count: 1
cpu-frequency: 2400MHz
cpu-load: 0%
free-hdd-space: 64.1MiB
total-hdd-space: 95.2MiB
write-sect-since-reboot: 2224
write-sect-total: 2225
architecture-name: x86_64
board-name: CHR
platform: MikroTik
Trying to import a 245MB Dude export obviously fails because there is not enough storage. Thoughts/Fixes?
This is running on GNS3 I can see the tags by using Wireshark, so the issue is not a lack of support.What virt environment? Some don't support vlans...
did you mean 'use-ip-firewall' in bridge settings? it should not affect Bridge FilterBridge firewall has to be enabled explicitly. It is off by default.
Same to me,iam getting this here
[raffaello@CHR-1] > system license renew account=raffav level=p1
password: ******
status: ERROR: This System ID with old ID already received trial
license key pause]
[raffaello@CHR-1] > system license print
system-id: Bfj8Y2IKdRJ
level: free
[raffaello@CHR-1] > system license generate-new-id
Your license will become invalid. Continue, yes? [y/N]:
y
[raffaello@CHR-1] > system license print
system-id: CCpSFPsCZkE
level: free
[raffaello@CHR-1] >
[raffaello@CHR-1] > system license renew account=raffav level=p1
password: ******
status: ERROR: This System ID with old ID already received trial
license key pause]
[raffaello@CHR-1] >
/system resource> print
uptime: 9m56s
version: 6.35rc8 (testing)
build-time: Feb/11/2016 07:07:30
free-memory: 219.7MiB
total-memory: 245.8MiB
cpu: Intel(R)
cpu-count: 4
cpu-frequency: 2394MHz
cpu-load: 0%
free-hdd-space: 14.8GiB
total-hdd-space: 14.9GiB
write-sect-since-reboot: 1176
write-sect-total: 1177
architecture-name: x86_64
board-name: CHR
platform: MikroTik
updateI am no longer able to update the CHR,
/system license> print
system-id: -
level: p1
limited-upgrades:yes
next-renewal-at: feb/21/2016 19:03:55
deadline-at: feb/16/2016 21:59:59
How do you get CHR to work on Xenserver . Still be trying with no luck .Hi guys,
Got it running on our XenServer 6.5 environment (CHR 6.34rc41), however, when I try to 'migrate' the vm to a different host it freezes. Also shutdown via Xen doesn't work (restart does).
Would it be possible to give the XenServer host some info via the Kernel? It currently looks like this:
I know that Xen support is possible via the Linux kernel because Debian/Ubuntu works (says Virtualisation state 'Optimised' and OS name) even without Xen Tools.
On a positive note: got full Gigabit performance between a Routerboard and CHR with the built-in speedtest. Nice.
I'm really looking forward to implement CHR in our environment. Starting as secondary router initially but who knows...
While I haven't played with VLANs myself, I've noticed that GNS3 switches have the option to specify a VLAN at each port, and there's no "pass-through" option - you must specify a VLAN, and it defaults to 1.is GNS3 by itself aware of the VLANs sometimes these virtual systems just bork the VLAN tags.
Same problem here except with trial license. Did you ever get it working?I bought a p1 license ,but the CHR's license is still free,how can i make it work ?
i tried to renew and also restarted the CHR,but it is still free
[admin@MikroTik] > /system license renew
account: ***
password: ************
level: p1
status: done
[admin@MikroTik] > /system license print
system-id: *****
level: free
[admin@MikroTik] >
i also sent email to support@mikrotik.com,but no response,
i am really in a hurry to fix this problem, pls reply soon
The GNS switches are VLAN aware if you change the port from access to dot1q. I've been using VLANs with CHR in GNS3 successfully since the CHR 6.32 imageis GNS3 by itself aware of the VLANs sometimes these virtual systems just bork the VLAN tags.
As per my earlier request I second that.just linking in my request http://forum.mikrotik.com/viewtopic.php?f=1&t=105277
I really would like to have this setup in HA, but need the integration tools/kernel module to do so. I believe this should be pretty easy to integrate, but can only be done by mikrotik.. Thanks.
Using the synthetic network adapter the boot loop occurs also on Windows 2012 R2 and Windows 8.1. With legacy network adapter the CHR boot correctly but the speed is limited to 100 Mbps.CHR 6.34rc41 (Release candidate) doesn't work with HYPER-V 1st generation. There is a boot loop after DSK key generation...
WIN 10 pro+hyper-v.
Using the synthetic network adapter the boot loop occurs also on Windows 2012 R2 and Windows 8.1. With legacy network adapter the CHR boot correctly but the speed is limited to 100 Mbps.CHR 6.34rc41 (Release candidate) doesn't work with HYPER-V 1st generation. There is a boot loop after DSK key generation...
WIN 10 pro+hyper-v.
On Windows 2012 (not R2) the CHR boots correctly also with synthetic network adapter. Probably it works also on Windows 8 (I didn't test this last configuration).
Could MikroTik help us to solve this problem? In our server farm all the hypervisors are 2012 R2 and we would like to buy some CHR licenses.
Thanks in Advace,
Davide
Bump.So stupid question time, unlimited license is still per interface or its unlimited?
I read this, and it leads me to believe the P1 and P10 is per interface and the P-Unlimited is just that, Unlimited. You pay $250 and Mikrotik doesn't care how many interfaces you have on your CHR and at what speed. I'm just trying to clear up my confusion as I don't want to tell my supervisor one thing and then have to go back and tell him something different later.
I was hoping to avoid the trial license because it will require a new one every time I spin up an instance.
HelloI don´t understand that 1Mbit limitation. It is only usefull to see if it runs or not.
For testing I would be much more happy with the 24h test-version (or even less time), to see how it performs compared to the "old" virtualized x86 version.
No, I am not. It would be cumbersome and not OK with systems that need to keep running. But possible it is.So you are advising to reload the CHR each 2 months and keep it running so for free for ages?
... the license.
I think he meant existing license.Thanks for being honest
Raffav, can you email support ?
Have you seen this:But I NEED CHR to shut down when the host tells it to
Version 6.35rc42 has been released.
Changes since previous version:
*) chr - fixed bridge firewall;
*) chr - make shutdown request from hyper-v work (might fix other hypervisor as well);
*) dude - fixed dude login logging, no more shows as winbox login;
*) timezone - fixed reboot by watchdog when selecting timezones from the end of list
I'm getting same error here....tried everything and keep getting this error
When installing routerOS..... almost done and it fails with same error:
The error is while installing the OS (from the download image)
What I've tried:
Install OS from VMware Workstation: Works.
Resize disk with VMware Workstation: Works
Migrate resized and installed image to ESXi: Work for sometime but after disk write it fails with same error
Download image and resize the RAW image with qemu: Resize Works
Convert RAW image to VMDK with qemu: Works
Create a Virtual Machine inside ESXi with the resized vmdk image by qemu: Works
Start virtual Machine on ESXi and start install: works
Finished install: SAME ERROR!
What is wrong ??!?!?!
HELP PLEASE!
Had you tryed the last 35.rc42?I'm getting same error here....tried everything and keep getting this error
When installing routerOS..... almost done and it fails with same error:
The error is while installing the OS (from the download image)
What I've tried:
Install OS from VMware Workstation: Works.
Resize disk with VMware Workstation: Works
Migrate resized and installed image to ESXi: Work for sometime but after disk write it fails with same error
Download image and resize the RAW image with qemu: Resize Works
Convert RAW image to VMDK with qemu: Works
Create a Virtual Machine inside ESXi with the resized vmdk image by qemu: Works
Start virtual Machine on ESXi and start install: works
Finished install: SAME ERROR!
What is wrong ??!?!?!
HELP PLEASE!
Nice day,
I discovered the same problem, I had to first use VMware Workstation for the installation VMDK, and then convert into ESXi by using:https://my.vmware.com/web/vmware/info/s ... dalone/6_1
Unfortunately it is not possible to select another ethernet card than the emulated E1000. VMware VMXNET3 card is not in the selection.
VMware Tools also do not work
Had you tryed the last 35.rc42?
Ps you will need winbox3.4 to access it
Hello Normis,VMWare tools are for Windows and such operating systems. RouterOS will not have it. All the needed network and storage drivers are already there, VMtools will not be there
Sob, thank you! Solved! I must change recommended "Other 64-bit" (http://wiki.mikrotik.com/wiki/Manual:CH ... stallation - step 2) to "Other 2.6.x Linux (64bit)".No need to request already included features, both vmxnet3 and pvscsi are already there, just select them when creating new VM.
Hello Normis,VMWare tools are for Windows and such operating systems. RouterOS will not have it. All the needed network and storage drivers are already there, VMtools will not be there
I understand that Mikrotik does not contain the complete VMware Tools, but should have implemented Linux drivers , at least for the network card: https://kb.vmware.com/selfservice/micro ... Id=1001805. It would be nice to have a SCSI controller for hard disks and as a bonus, heartbeat function. But then you have to support only one version of VMware Solution: Workstation or better ESXi (or ESX paid).
Without the support Linux drivers not VMDK image continues to make sense.
Hello,Hello,
There is an issue with the ethernet interface order on CHR using VMware (tested on EXSi and Workstation). When you add several interfaces to the CHR then the interfaces appear in different order as created. This means they cannot be matched easily when connecting to other VMs. You have to check the MAC address of the interfaces to match then correctly.
Here is a supout file:
https://dl.dropboxusercontent.com/u/381 ... ut-chr.rif
Here is an example:
All the interfaces are "VMXNET 3"
I tested with stable and last RC and the same happened.
This issue is not present on x86 version.
Best regards.
Please expand on this disk resize for CHR. I tried on a QEMU virtual server but after expanding the router wouldn't reboot, it said unable to find operating system. So I restored a backup and the CHR restarted with a disk size of 128MB on a 1GB partition. I'm currently running 6.35rc48.on x86 (non-chr) it formats the disk and tries to use entire disk when installing. CHR is for virtual machines and the assumption is that it is running on the virtual disk, that can be expanded if you so desire.
if CHR sees that disk suddenly has expanded, it will attempt to expand the storage partition. On CHR installation that was done with the version older than 6.35rc36 this will not work as additional partition table changes were needed for this to work at all
All good now. Original version was RouterOS 6.35rc35 then RouterOS 6.35rc42 and finally 6.35rc48. I've reformatted and rebuilt it with RouterOS 6.35rc48, restored the configuration and expanded the OS partition successfully.CHR guest has to be made using RouterOS 6.35rc42 or newer. Note the __made__. If it is made with version, then resize will work. If this was made with older - it will not, even if you update to the latest version.
Any news/instuctions about this topic?I will be great if we can do some type of migration license from x86 to CHR since 3 type of level license has the same price
Level 4 = P1 ($45)
Level 5 = P10 ($95)
Level 6 = Unlimited ($250)
Hi FractionAny news/instuctions about this topic?I will be great if we can do some type of migration license from x86 to CHR since 3 type of level license has the same price
Level 4 = P1 ($45)
Level 5 = P10 ($95)
Level 6 = Unlimited ($250)
I saw Normis asked raffav to contact support after this post, is that the way to get the license migrated?
Hello, Raffaello.
Thank you very much for your idea.
We do not have plans to make any migration at the moment.
Regards,
Sergejs
This is fantasticit is easier now to make Amazon instances of the CHR, no more images to move around clusters:
https://aws.amazon.com/marketplace/pp/B01E00PU50
http://forum.mikrotik.com/viewtopic.php ... 31#p517531I can live-migrate... but without the xentools kernel module installed, the CHR is freezing on resume. I've found plenty of posts about the linux kernel freezing the same way then xentools isnt installed. I can't pull anything of value out of the CHR's logs so can only assume this is the issue.
I tried today with 6.36rc12. Installing/setting up a VM works flawless, but live migration still does not work as expected. XenCenter shows the VM on the new host, but the network stops working, all vCPU went to 100% and console stops responding. So there is no chance to watch what is going on in VM. Migrating the VM back to the origin host does not work either. Although XenCenter says first migration completed, it seems like the migration did not. XenCenter shows migration error for second migration.Got it running on our XenServer 6.5 environment (CHR 6.34rc41), however, when I try to 'migrate' the vm to a different host it freezes. Also shutdown via Xen doesn't work (restart does).
Waoo That's Great i will definitely try this but i just say one thing to you if you want to add some more things on that and new feature you can get some cool instruction from that.We are releasing a test version of an exciting new feature - Cloud Hosted Router (CHR).
It is a Virtual Machine image of RouterOS that has full functionality of RouterOS without any kind of conventional RouterOS license, with the limitation of 1Mbit per interface, in future, we will offer unlimited speed with a paid subscription. This allows you to use RouterOS in training classes, for testing, for experimentation or any other purpose, without the need for 24h trial or Demo license.
We provide a generic (RAW) format IMG, VMDK file for VMWare, VDI for VirtualBox and VHDX for Hyper-V. These are ready to use virtual disk images (add as a virtual hard drive to your VM). These are NOT templates for importing. Cdn reviews
qemu commands for converting images (if you need to convert to another format)
RAW to VMware:RAW to VirtualBox:Code: Select allqemu-img convert -f raw -O vmdk chr-6.31rc9.img chr-6.31rc9.vmdk
See below for steps how to run this new system.Code: Select allqemu-img convert -f raw -O vdi chr-6.31rc9.img chr-6.31rc9.vdi
If you need a bigger virtual hdd (currently 128Mb), resize the image before first runThis will increase the disk by 1GBCode: Select allqemu-img resize -f raw chr-6.31rc9.img +1G
Downloads:
http://www.mikrotik.com/download
VMware Player (Free): https://my.vmware.com/web/vmware/free#d ... player/7_0
VirtualBox (Free): https://www.virtualbox.org/wiki/Downloads
Instructions for VMware:
Screenshot 2015-07-27 14.44.27.png
Screenshot 2015-07-27 14.44.31.png
Screenshot 2015-07-27 14.44.34.png
Screenshot 2015-07-27 14.44.39.png
Screenshot 2015-07-27 14.44.46.png
Screenshot 2015-07-27 14.44.50.png
Screenshot 2015-07-27 14.44.57.png
Screenshot 2015-07-27 14.45.06.png
Screenshot 2015-07-27 14.45.20.png
Instructions for VirtualBox:
Screenshot 2015-07-27 14.46.09.png
Screenshot 2016-03-11 14.10.24.png
Screenshot 2016-03-11 14.10.35.png
Screenshot 2015-07-27 14.47.16.png
Screenshot 2015-07-27 14.46.28.png
IMPORTANT STEP, change CPU setting to KVM:
Screenshot 2015-07-27 14.46.51.png
Screenshot 2015-07-27 14.46.55.png
Screenshot 2015-07-27 14.47.16.png
Probably you could try clonezilla:How to deploy CHR on KVM VPS with Virtualizor Client panel ?
the provider Just support ISO instalation file ?
one of the most important reasons to have vm-tools is so that the hyper-visor is aware of the state of the VM, and can reboot it if no heartbeats are received. Check out https://pubs.vmware.com/vsphere-50/inde ... D78E7.htmlVMWare tools are for Windows and such operating systems. RouterOS will not have it. All the needed network and storage drivers are already there, VMtools will not be there
HiHello,
Here is an OVA optimized for VMware vSphere with SCSI hard disk and SCSI controller (paravirtualization). The hard drive has 1GB.
Version: 6.35.2
OVA: https://dl.dropboxusercontent.com/u/381 ... phere6.zip
It can be used on VMware Workstation as well.
Enjoy!
Best regards.
Agreed. In Cluster environments (SAN etc) SCSI is a kind of a must, because IDE over iSCSI is just terrible. I think having proper support for ESXi is the only path. VMWare Workstation - is not for production enviroment, its for testing enviroments.Hello Normis,VMWare tools are for Windows and such operating systems. RouterOS will not have it. All the needed network and storage drivers are already there, VMtools will not be there
I understand that Mikrotik does not contain the complete VMware Tools, but should have implemented Linux drivers , at least for the network card: https://kb.vmware.com/selfservice/micro ... Id=1001805. It would be nice to have a SCSI controller for hard disks and as a bonus, heartbeat function. But then you have to support only one version of VMware Solution: Workstation or better ESXi (or ESX paid).
Without the support Linux drivers not VMDK image continues to make sense.
cd /root
curl -O http://download2.mikrotik.com/routeros/6.35.2/chr-6.35.2.img.zip
gunzip -S .zip chr-6.35.2.img.zip
umount /mnt/vdb1
dd if=/root/chr-6.35.2.img of=/dev/vdb
RouterOS does not need graceful shutdown. OK, except maybe if you added some config and need to turn off immediately.Hi there,
I am a potential customer, the potential comes from the lack of open-vm-tools. Being unable to gracefully shutdown/restart the vm without logging on is a no-go for me. Once you add that you have a at least a p1 license sold
Is router os running from a read-only filesystem? If it doesn't, there is a real risk about the FS getting corrupted if i "yank the power cord", shutting it down without the agents, everytime i service my host. I understand about keeping the attack surface as slim as possible but adding the open-vm-tools package as an optional pkg shouldn't be much work, the code is opensource(https://github.com/vmware/open-vm-tools) so you just need to compile it.RouterOS does not need graceful shutdown. OK, except maybe if you added some config and need to turn off immediately.
RouterOS might not need it but the virtualization host totally depends on it. No proper hypervisor will kill a VM when it signals all VM to shutdown normally. Killing a VM, as you suggest, is a totally different action than shutting it down!RouterOS does not need graceful shutdown. OK, except maybe if you added some config and need to turn off immediately.
Good to know then, another thing that having the tools adds is the fact that my standalone hosts autoloads a series of vms and having the agents working cuts dramatically the time before the next autostart vms starts(instead of waiting 30 minutes it will instantly startup the next in order). Logically i would load the router before the rest of the stack and adding say 15 to 30 minutes for each reboot is something i would prefer to avoid if i can.I have never seen a FS problem and everyone here is "yanking the cord" for 15 years or so.
Yeah, the my main issue with the lack of agents is that, without the agents, i will need to have the host wait for the vm delay at every host power up(which is a cool 15 minutes for my homelab) that would be zero if the agents were thereJust checked the free version of ESXi, it allows to configure shutdown action (with options "Guest Shutdown", "Power Off" and "Suspend" available) and startup and shutdown delays on a per VM basis. I'm pretty sure any decent virtualization platform should allow to set such options as well.
This worked like a charm... beautifully! However, I used another VPS provider . Similar conceptIf you are interested IN CHR on OVH VPS SSD, here you go:
1. Install any linux distro
2. From OVH control panel boot VPS to rescue mode
3. Type the magic from below :Code: Select allcd /root curl -O http://download2.mikrotik.com/routeros/6.35.2/chr-6.35.2.img.zip gunzip -S .zip chr-6.35.2.img.zip umount /mnt/vdb1 dd if=/root/chr-6.35.2.img of=/dev/vdb
I belive that quantity of ethernet interfaces is provider-depended thing.This worked like a charm... beautifully! However, I used another VPS provider . Similar conceptIf you are interested IN CHR on OVH VPS SSD, here you go:
1. Install any linux distro
2. From OVH control panel boot VPS to rescue mode
3. Type the magic from below :Code: Select allcd /root curl -O http://download2.mikrotik.com/routeros/6.35.2/chr-6.35.2.img.zip gunzip -S .zip chr-6.35.2.img.zip umount /mnt/vdb1 dd if=/root/chr-6.35.2.img of=/dev/vdb
Are you able to apply more than one(1) ethernet interface? I only have one(1) by default. Probably can just create VLAN's if needed...
Yes there is console screen, but read only. Again, this is Amazon limitation. Interactive console is not available for any OS in Amazon EC2: http://docs.aws.amazon.com/AWSEC2/lates ... nsole.htmlWhen you will do the CHR version for Amazon and Azure please notice that there is not support of remote display and console. The only way to make it accessable is to introduce to network infrastruture and create a real working profile for current routerOS such as Home AP and etc.
The license as far a I know is for instance, so you need to buy 5 license but you have the possibility to move the license to another instance when neededHi Normis,
Congratulations to you and your team for a solid product. The CHR on KVM seems to be working nicely so far.
Quick question: Let's say we have 5 powerful bare metal servers spread across continental US. And we have 5 total instances of "CHR + KVM" running on each of those five servers forming a secure IPSec circuit.
Will I need to purchase 5 separate PI-10 licenses to allow those five instances to work normally at 10Gigbit throughput? Or does buying one "PI-10 unlimited" perpetual license allows me to apply it to multiple CHR instances?
Kindly please help advice at the earliest.
-Amit
The CHR license I think only is possible to purchase on mikrotik site via your account also there is where you manage your CHR licenseRaffav and Normis, thank you for the clarification.
Normis, as far as the ability to move the license around between different CHR instances is concerned: Is this still valid if I were to purchase your CHR PI-10 licenses from a reseller? Does it have to be any special type of PI-10 licenses? Or does buying regular CHR PI-10 in bulk good enough.
Also - What specific steps should we have to follow for moving the currently active license key to a new CHR instance? Let's assume I had licensed CHR running and I just deleted that instance. Now I need to re-purpose that license key to a new CHR instance. The RouterOS does not allow me to change its router identity key.
best
-Amit
You can't buy CHR licenses from a reseller. What you can do, is buy a "coupon" (a prepaid key) and then use that to pay for your CHR. But in reality this still means you make your own CHR keys.Normis, as far as the ability to move the license around between different CHR instances is concerned: Is this still valid if I were to purchase your CHR PI-10 licenses from a reseller?