Slow flow on TIV

Recently, I have configured a VTI interface between two 2921 routers.  The link between the two routers is 100 MB, but the tunnel does not seem to be able to spend a lot more traffic.  I noticed a few differences between the MTU and bandwidth in the Tunnel interface compared to the physical interface.  Is it normal for a VTI?

Tunnel1 is up, line protocol is up

Material is Tunnel

The Internet address is 192.168.193.127/31

MTU 17862 bytes, BW 100 Kbps, DLY 50000 usec,

reliability 255/255, txload 43/255, rxload 99/255

Encapsulation TUNNEL, loopback not set

KeepAlive not set

Tunnel source 1 *. ***. ***. 1, destination 1 *. ***. ***. 2

Transport/Protocol of IP/IPSEC tunnel

TTL 255 tunnel

Tunnel transport MTU 1422 bytes

Tunnel of transmission bandwidth 8000 (Kbps)

Tunnel to receive 8000 (Kbps) bandwidth

Tunnel of protection through IPSec (profile "RD_VTI")

Last entry 4w2d, exit 4w6d, blocking exit ever

Last clearing of "show interface" counters 4w2d

Input queue: 0/75/0/0 (size/max/drops/dumps); Total output drops: 176

Strategy of queues: fifo

Output queue: 0/0 (size/max)

30 second entry rate 123000 bps, 157 packets/s

exit rate of 30 seconds 1994000 bps, 233 packets/s

836701737 package, 47577560492 bytes, 0 no buffer entry

Received 0 emissions (0 of IP multicasts)

0 Runts, 0 giants, 0 shifters

errors entry 0, 0 CRC, overgrown plot of 0, 0, 0 ignored, 0 abort

1506226971 packets output, 1926214877370 bytes, 0 underruns

0 output errors, 0 collisions, 0 resets interface

unknown protocol 0 drops

0 output buffer failures, 0 output buffers swapped out 30 second entry rate 123000 bps, 157 packets/sec

Disclaimer

The author of this announcement offers the information in this publication without compensation and with the understanding of the reader that there is no implicit or explicit adequacy or adaptation to any purpose.  Information provided is for information purposes only and should not be interpreted as making the professional advice of any kind. Use information from this announcement is only at risk of the reader.

RESPONSIBILITY

Any author will be responsible for any damage that it (including, without limitation, damages for loss of use, data or profits) arising out of the use or inability to use the information in the view even if author has been advised of the possibility of such damages.

Poster

Without a thorough analysis, your adjust-mss seems too small; IP MTU minus 40 is the commune.

Benefits activation of PMTUD on the tunnel interface (see my previous Cisco white paper reference).

So, what does the mower to the shape of the 95% to?  Why are shape you?

Tags: Cisco Network

Similar Questions

  • EA6900 Super slow flow

    I have an EA6900 put in place for the routing functions on a Virgin Media 120Mbps cable connection.  I see a very slow flow on the Linksys router, and after looking for obvious causes, I am stumpted.

    The EA6900 is connected to Virgin Media "Superhub" which is on the modem mode.  In this configuration, a single peripheral cable with a gigabit ethernet card can see download speeds autour 9Mbps.  However, when a Superhb in router mode switch and connect a device directly to it, jump to 100 Mbit/s download speeds.

    I tried several different network between the modem and the EA6900 cables, but the results are the same.  I was also in touch with Virgin Media, who said that there is no problem with the upstream connection.

    He suggests that the cause is something to do with the EA6900.  Any suggestions would be greatly appreciated.

    Ben

    Disable media Prioritizaion or leave it turned on and ensure that you set the speed of bandwidth downstream to 120000 in advanced settings. Let us know if that helps.

  • SonicWall NSA 220 slow flow

    Hi all

    According to specification of 220 SonicWall NSA could handle a flow rate of 600 Mbps without security options and the only inspection of the firewall.

    We have recently acquired a 500 Mbps (up and down) line, so it was time to put this NSA220 to the test. And it has failed.

    The maximum through put has 175 top down and tried many things. I also tried by default with https://support.software.dell.com/kb/sw8119 of the instructions on how to get the throughput, but always to the max of 175 Mbps. Again, I disable all security options and their disabled in the configuration of the area.

    So I tried the following thing:

    1. Put my computer in connection to a laptop and start the transfer a SMB file. The result is a speed of 900 Mbit/sec.

    2. Put my computer on the local network of the SonicWall and the laptop on the Wan and tried again the transfer of the laptop (WAN) to the computer (LAN). Result: 175 Mbit/max.

    What am I missing or doing wrong?

    Please refer to the post
    http://en.community.Dell.com/TechCenter/security/network-mobile-email/f/4904/p/19610851/20825216#20825216

    I have provided an explanation on a similar question it.

  • Slow flow on MPLS VPN WAN

    Anyone have any ideas why a portion of the traffic is slow as it passes through a VPN MPLS WAN. My FTP copies are fast but copy all windows or windows file transfers are slow. Copies of windows are about three times slower as the FTP transfers. Can be optimized on routers or switches?

    Hello

    Thus, all transfers are done with CIFS are slow and other then CIFS are ok?

    All transfers are between XP/7 and servers (before 2008)?

    Please take a look at http://bit.ly/rkh9IM

    CIFS (or SMB) prior to the 2008 version is slow by definition as it can not cope with very good latency. Other protocols such as HTTP and FTP run much smoother.

    When you run Server 2008 (or better) combination with Windows Vista (or better) should solve some of your problems as it can using SMBv2.

    What actual speed is your order on the MPLS and what is the maximum transfer reached between server and workstation?

    Best regards, G.

  • Domestic network file share is slow

    Hello

    I'm having a problem with a slow flow on my LAN, my main office and another office in a nearby room, which is used as a digital recorder. Both are fine Internet download about 2.0Mbps + effective flow. But the copy of the files between them (which should be a lot faster!), I'm moving 400 to 500 Kbps. Since one is used as a digital video recorder, it is a problem, try to copy large video files, or stream videos in HD quality on my network who stutters. It's frustrating getting the best quality watch Netflix, than from another computer in my own home.

    Both have compatible 802.11n wireless NIC; one is a card PCI Rosewill RNX-N300, and the other is a TP-LINK WN951N PCI - E card. I use the Atheros drivers on the machine of TP-LINK; on the other hand, it seems that Atheros does not support the Rosewill cards very well, so it is the driver of stock. Even on the TP-LINK machine; using the factory instead of Atheros driver, doesn't seem to help. My router is configured for WPA2 802.11n 'Mode of Performance '. I also tried Compatibility Mode and everything else I can think.

    I realize that the Atheros drivers are not supported here, but I don't think it's relevant, as the factory, or native Windows 7 generic NIC drivers appear to have the same problems. I tried to minimize interference with cards, which usually get 3-4 of 5 bars. WAN speed being much faster, I doubt that's the issue. Wired ethernet is out of the question for me, but pretty much the same speed powerline adapters, seems no better/worse. LAN ping times are 1-2 ms.

    Is there something I'm missing here? Both machines are running Windows 7 Pro x 64. I wonder if there is a key "reg" to the packet size or anything else that might help. I would chalk up to the nature of wireless cards, if it wasn't for much better WAN speeds. I believe that it is a problem of operating system settings; all other devices on the network, continue to get faster internet blazing ascent/descent speeds, even though the slow file copies/streaming are underway.

    Thanks for your suggestions.

    Hello

    This might help:

    http://www.SevenForums.com/network-sharing/218886-very-slow-network-file-sharing.html

    http://www.SevenForums.com/network-sharing/25843-slow-file-transfer-over-network.html

  • Your movie will be ready to watch in 9 hours and 43 minutes.

    I thought that with the advent of the new Apple TV that this problem would be resolved or taken more seriously. Maybe with the rumors that they will create ther own servers for the cloud, it will be, but at the same time other companies continuously seem to be able to exploit this fundamental requirement without problem: I can not watch so often not what I bought when I want to. Last night for example, I went to watch an episode of television and was greeted with a timeout of 9 hours and 43 minutes. Yes, it does improve not only after starting to watch Hulu or Netflix or similar and normally when I'm halfway through watching something else. This problem also extends to the trailers. Sometimes trailers take as long as the trailer to start to play and still in YouTube it will begin instantly. Why? It's infuriating.

    It's particularly frustrating because she is content you own. Is there a solution to this? other than to submit comments is there a way to get this question considered a greater priority? I'm sure this is but lights.

    What version of ATV, which software?

    What is your internet connection speed?

    What the ATV that is connected to the network?

    It's widely reported that iTunes always transmits movies in full 1080 p. Other services, particularly from Netflix, use a lower resolution when they detect a slower flow. For more information, see get help play or content streamed to your Apple TV - Apple Support .

  • Stream 14 bad performance - driver installation order

    I use a 14 - sound incredibly slow flow.

    I realize it's a low-end PC range, but if I open more than 4 or 5 tabs it begins to freeze.

    I tried to reinstall, maybe the drivers are not presented properly.

    Is this normal for this PC?

    I have this list to install, but not able to find the correct drivers, as they call the same.

    What is the Chipset and ethernet called on the site - they are under another name on the site.

    Thank you

    On any PC, you must:

    1. install the chipset drivers and restart.

    2. install the graphics driver and restart.

    3. install the audio driver and restarted.

    4 install the ethernet driver.

    5. install the driver for the wireless card.

    Then, you can install any other drivers you need in no particular order.

    I also found I think that's the right bluetooth driver you need...

    FTP://ftp.HP.com/pub/SoftPaq/sp70501-71000/sp70725.exe

  • CAN order on not always transmitted XNET

    I try to send a message BOX using a PXI-8512.  I have an updated database in place.  When I write CAN data, it turns on an icon on a screen.   It works 90% of the time on a known good unit.  However, 10% of the time, I call this sub - vi (below) which is supposed to enable the icon, and the icon never lights.  I thought that maybe I wasn't give him enough time before I closed the session, so I increased the timer loop to 1500 ms, but that did not help.  (I think the timer loop was 500 ms before I increased it).  Does anyone have any ideas as to why this message is not received correctly some of the time?

    The timer is running, but it might be the first or last, as you don't sequence them at all.

    The CAN bus is a shared bus. 1.5 s between the images is probably long enough, but if you use a slow flow and the bus is busy, you might get squeezed out.

    Your other option, if you want all images to convey, is to use a session in queue instead of a single point of the session. And you expect to send to be complete prior to the closing.

  • RE1000 stuck in 802. 11 b / g mode of

    Just bought this camera today and very disappointed, as I've been with several of my recent purchases of Linksys.

    In any case, try to make the best of it... after hours of messing around to configure and test, I found that the RE-1000 is in 802. 11 b / g mode, whereas my router WAG120N presents itself as the 802.11n (it is actually being run in mixed mode for g and n devices)

    I wonder if anyone knows how I can get out of 802. 11 b / g mode and somehow get it up to speeds n we have n devices connecting to the RE1000 and get a very slow flow.

    Any help or advice would be greatly appreciated!

    As the work of the range extender is to extend the signal wireless to the router, it should be able to pass the signal on mode N also. Try setting the security your gateway to WPA2 wireless if it is not already set to that. You may also need to change the channel of your gateway wireless. Try to define the Standard 20 MHz on your gateway radio band and use channel 1, 3, 6, 9, or 11. Check the firmware version on the range extender as well because he might be running on the old version. You can check the version of the firmware here: http://support.linksys.com/en-eu/support/rangeexpanders/RE1000.

  • Green Radio 371 WAP name and Ethernet

    I just bought 3 WAP371 for a busy medical practice (course of the DR 6 from exam room to the examination room with laptops)

    I intend to put up with some Configurations of Point unique.

    Can I name Radio 1 and 2 of the Radio of the same name? Also, can the two share the same WPA2/AES password?

    Finally, going green Ethernet mode enabled because of the problems of performance or hours of slow connection?

    Thank you

    JB

    Hello [email protected] / * /;

    To answer your questions

    Can I name Radio 1 and 2 of the Radio of the same name? Also, can the two share the same WPA2/AES password?

    Answer: You can, but I usually recommend having backup under another name, so if they're going to use the radio as their primary 5.0, they can connect to this one.  If they start to have a slow flow due to the range or according to what others are downloading download, they can connect to radio 2.4.  Security you can keep the same if you wish.

    Going green Ethernet mode enabled because of the problems of performance or hours of slow connection?  I didn't know this to be a problem, but if it's a problem you can disable it.

    Kind regards

    Michael D.

    If this post is helpful please note or mark it as correct.

  • No 100 Mbps with cisco 892 router, cpu caps

    Hello

    We have a connection internet-based corporate (100 Mbit/s down, 10 Mbps upward, certain guarantees) with eight fixed ip addresses (in the a.79 configuration.72.  To join our network to the internet, we use a router 892.

    I am new to cisco equipment, so I had some trouble getting things to work, but I arrived.

    Now, we are facing this problem of throttling: whenever we are using more about 60mbit/s down bandwidth, this router CPU is maxing out (98-99% with 'show processes cpu history').

    When I download a torrent (dvd debian) to 4 MB/s (or 48mbit/s) cpu running at about 46 percent (tops at 49%).  Stop the download of results in 14% at the top of the CPU usage.

    When you use the command 'See deals cpu sort', I get this:

    Maximilian #show process cpu sort

    CPU utilization for five seconds: 46% / 43%; 01:00 %; 05:00 %

    Process PID Runtime (ms) Invoked uSecs 5 Sec 1 Min 5 Min TTY

    82 35978164 10389766 3462 2.31% 2.08% 2.06% 0 COLLECT NECK STAT

    90 203264 266300211 0% 0.31 0.31% 0.28% Ethernet 0 Msec Ti

    31 664 844 786 0.31% 0.19% 0.08% 8 SSH process

    108 2039472 5100352 399 0.23% 0.34% 0.30% IP 0 comments

    334 35468 4269539 8 0.23% 0.09% 0.04% 0 IP NAT Ager

    324 21204 2077221 10 0.07% 0.03% 0.02% 0 jobs per second

    104 24240 64783802 0 0.07% 0.04% 0.02% 0 IPAM Manager

    336 7404 108003 68 0.07% 0.02% 0.00% 0 IP VFR proc

    33 66952 321851 208 0.07% 0.00% 0.00% ARP 0 comments

    9 0 2 0 0.00% 0.00% 0.00% 0 timers

    ...

    If the CPU usage is about 46%, so that no process uses actually more than 2.31%.  In addition, these numbers do not change if I stop the download.

    It's our configuration (with parts obscured):

    Maximilian #show run

    Building configuration...

    Current configuration: 8035 bytes

    !

    ! Last configuration change at 09:49:27 UTC Wednesday, May 30, 2012 by jan

    ! NVRAM config update at 13:55:40 UTC Tuesday, May 22, 2012 by jan

    ! NVRAM config update at 13:55:40 UTC Tuesday, May 22, 2012 by jan

    version 15.2

    horodateurs service debug datetime msec

    Log service timestamps datetime msec

    no password encryption service

    !

    maximilian hostname

    !

    boot-start-marker

    Start the flash config: maxi-config

    boot-end-marker

    !

    !

    logging buffered 51200 warnings

    !

    No aaa new-model

    !

    Crypto pki token removal timeout default 0

    !

    Crypto pki trustpoint TP-self-signed-3260749506

    enrollment selfsigned

    name of the object cn = IOS - Self - signed - certificate - 3260749506

    revocation checking no

    rsakeypair TP-self-signed-3260749506

    !

    Crypto pki trustpoint tti

    crl revocation checking

    !

    !

    TP-self-signed-3260749506 crypto pki certificate chain

    certificate self-signed 01

    ........... [snip]...

    quit smoking

    encryption pki certificate chain tti

    !

    !

    !

    DHCP excluded-address IP 10.10.10.1

    IP dhcp excluded-address 192.168.1.0 192.168.1.49

    !

    CVO-IP dhcp pool pool

    import all

    Network 10.10.10.0 255.255.255.248

    default router 10.10.10.1

    Server DNS 8.8.8.8

    Rental 2 0

    !

    Maxi-pool of IP dhcp pool

    import all

    network 192.168.1.0 255.255.255.0

    default router 192.168.1.1

    Server DNS 8.8.8.8

    Infinite rental

    !

    !

    IP domain name adhese.org

    8.8.8.8 IP name-server

    inspect the IP name DEFAULT100 ftp

    inspect the IP h323 DEFAULT100 name

    inspect the IP icmp DEFAULT100 name

    inspect the IP name DEFAULT100 netshow

    inspect the IP rcmd DEFAULT100 name

    inspect the IP name DEFAULT100 realaudio

    inspect the name DEFAULT100 rtsp IP

    inspect the IP name DEFAULT100 esmtp

    inspect the IP name DEFAULT100 sqlnet

    inspect the name DEFAULT100 streamworks IP

    inspect the name DEFAULT100 tftp IP

    inspect the tcp IP DEFAULT100 name

    inspect the IP udp DEFAULT100 name

    inspect the name DEFAULT100 vdolive IP

    IP cef

    No ipv6 cef

    !

    !

    !

    !

    Authenticated MultiLink bundle-name Panel

    !

    !

    !

    !

    !

    !

    license udi pid CISCO892-K9 sn FC [snip]

    !

    !

    username secret privilege 15 cisco 5 [snip]

    username privilege 15 jan

    !

    !

    !

    !

    !

    property intellectual ssh pubkey-string

    jan username

    ssh - rsa [snip] jan@[snip key hash]

    quit smoking

    !

    !

    !

    !

    !

    !

    !

    !

    !

    interface BRI0

    no ip address

    encapsulation hdlc

    Shutdown

    Multidrop ISDN endpoint

    !

    interface FastEthernet0

    no ip address

    spanning tree portfast

    !

    interface FastEthernet1

    no ip address

    spanning tree portfast

    !

    interface FastEthernet2

    switchport access vlan 2

    no ip address

    spanning tree portfast

    !

    interface FastEthernet3

    no ip address

    spanning tree portfast

    !

    interface FastEthernet4

    no ip address

    spanning tree portfast

    !

    interface FastEthernet5

    no ip address

    spanning tree portfast

    !

    FastEthernet6 interface

    switchport access vlan 2

    no ip address

    spanning tree portfast

    !

    interface FastEthernet7

    switchport access vlan 2

    no ip address

    spanning tree portfast

    !

    interface FastEthernet8

    IP 192.168.3.2 255.255.255.0

    automatic duplex

    automatic speed

    !

    interface GigabitEthernet0

    [snip] 255.255.255.252.94 IP address

    IP access-group 101 in

    no ip redirection

    no ip unreachable

    no ip proxy-arp

    penetration of the IP stream

    NAT outside IP

    inspect the DEFAULT100 over IP

    IP virtual-reassembly in

    automatic duplex

    automatic speed

    !

    interface Vlan1

    IP 10.10.10.1 255.255.255.248

    IP access-group 100 to

    IP nat inside

    IP virtual-reassembly in

    IP tcp adjust-mss 1452

    !

    interface Vlan2

    IP 192.168.1.1 255.255.255.0

    penetration of the IP stream

    IP nat inside

    IP virtual-reassembly in

    !

    IP forward-Protocol ND

    !

    IP fragment offset stream capture

    IP-length of stream capture package

    TTL for IP stream capture

    capture IP stream vlan id

    ICMP IP stream capture

    IP ip id stream capture

    IP stream capture mac addresses

    !

    IP http server

    23 class IP http access

    local IP http authentication

    IP http secure server

    IP http timeout policy slowed down 60 life 86400 request 10000

    IP nat pool system3-74 [snip].74 [snip].74 prefix length 29

    IP nat inside source list 74 pool system3-74 overload

    IP nat inside source static tcp 192.168.1.42 22 [snip] extensible 22.72

    IP nat inside source static udp 192.168.1.42 1194 [snip] extensible.72 1194

    IP nat inside source static tcp 192.168.1.42 22 [snip].72 extensible 1489

    IP route 0.0.0.0 0.0.0.0 [snip].93

    IP route 0.0.0.0 0.0.0.0 192.168.3.1 5

    IP route 10.8.0.0 255.255.255.0 192.168.1.42

    !

    access-list 1 permit 10.10.10.0 0.0.0.7

    access-list 23 allow 192.168.1.42

    access-list 23 allow 10.10.10.0 0.0.0.7

    access-list 74 allow 10.10.10.0 0.0.0.7

    access-list 74 permit 192.168.1.0 0.0.0.255

    access-list 100 deny ip 255.255.255.255 host everything

    access-list 100 deny ip 127.0.0.0 0.255.255.255 everything

    access ip-list 100 permit a whole

    access-list 101 deny ip 10.0.0.0 0.255.255.255 everything

    access-list 101 deny ip 172.16.0.0 0.15.255.255 all

    access-list 101 deny ip 192.168.0.0 0.0.255.255 everything

    access-list 101 deny ip 127.0.0.0 0.255.255.255 everything

    access-list 101 deny ip 255.255.255.255 host everything

    access-list 101 permit tcp any host [snip].72 eq 1489

    access-list 101 permit tcp any host [snip].72 eq 22

    access-list 101 permit udp any host [snip].72 eq 1194

    access-list 199 deny ip any host 74.209.133.138

    access ip-list 199 permit a whole

    not run cdp

    !

    !

    !

    !

    !

    SNMP-server [snip] RO community

    !

    control plan

    !

    !

    !

    !

    profile MGCP default

    !

    !

    !

    !

    exec banner ^ C

    % Warning of password expiration.

    -----------------------------------------------------------------------

    Virtual office of Cisco (CVO) is installed on this device and it provides the

    default name "cisco".

    It is strongly recommended that you create a new user name with a privilege level

    15 using the following command.

    username secret privilege 15 0

    Replace and with the username and password you want

    use.

    -----------------------------------------------------------------------

    ^ C

    connection of the banner ^ C

    -----------------------------------------------------------------------

    Virtual office of Cisco (CVO) is installed on this device and it provides the

    default name "cisco".

    It is strongly recommended that you create a new user name with a privilege level

    15 using the following command.

    username secret privilege 15 0

    Replace and with the username and password you want

    use.

    For more information about CVO, please go to http://www.cisco.com/go/cvo

    -----------------------------------------------------------------------

    ^ C

    !

    Line con 0

    local connection

    line to 0

    line vty 0 4

    access-class 23 in

    local connection

    length 0

    transport input telnet ssh

    line vty 5 15

    access-class 23 in

    privilege level 15

    local connection

    transport input telnet ssh

    !

    end

    Please tell me what I can do about it.  Or this router is not able to do 100 Mbps?

    The net effect of running out of CPU is random connections abandonment and no possible communication with the router.  I could cap the bandwidth 90mbit/s or 80mbit/s, but I'd rather not.

    The system of image files is: "flash: c890-universalk9 - mz.152 - 1.T1.bin.

    Thanks in advance!

    Jan.

    Disclaimer

    The author of this announcement offers the information in this publication without compensation and with the understanding of the reader that there is no implicit or explicit adequacy or adaptation to any purpose.  Information provided is for information purposes only and should not be interpreted as making the professional advice of any kind. Use information from this announcement is only at risk of the reader.

    RESPONSIBILITY

    Any author will be responsible for any damage that it (including, without limitation, damages for loss of use, data or profits) arising out of the use or inability to use the information in the view even if author has been advised of the possibility of such damages.

    Poster

    In fact, the 890 series is rated at 100 Kpps, i.e. approximately 51 Mbit/s (noted also in other posts) for minimum size Ethernet packets, but Cisco also documents the 890 providing up to 1 400 Mbit / s 1500 bytes of the packets of size.  Unfortunately "your mileage may vary." i.e. actual throughput is very dependent on your particular traffic and you configure your router to do against this traffic.  For example, you have NAT/PAT, ACLs, firewall inspection and NetFlow, all who consume extra CPU during the processing of packages of the interface.

    Not knowing what exactly will a customer with a router, Cisco makes recommendations to use very conservative, and for the 890, it recommends side WAN does not exceed 15 Mbps (duplex).  Again, it is very conservative, and as you have discovered, your configuration hit the wall about two times this recommendation, although it is unfortunately not enough to manage your bandwidth capacity.

    As other posters have noted, long-term or preferred solution is probably getting and using a faster router.  You can probably get out a title plus your 892 with additional 'tuning '.  That is, by eliminating all what you really, really need and do what you need as efficiently as possible.  For example, disabling NetFlow (such as already mentioned in some of the messages), disabling the firewall dynamic as you have NAT/PAT and ACL; and the "Resequencing" (if it is logically possible) ACEs.

    Regarding your question of the use of the services of police or release in the form, to avoid any overrun of the CPU (which really want to avoid!), Yes, something can be done there and could be very beneficial, but you will use some CPU for that and a really intelligent approach would be complicated.  (An example of 'smart' approach would be a built-in script that queries CPU frequently or traps these slow flows on high CPU, which then finds the high flows and policies dynamically.  A not-so-smart approach would be one policeman static for all incoming traffic, or only certain types of traffic entering the police.)

  • Performance problem with the guest operating system

    Hello gurus - I have one of the most unusual questions, I ran into a few time.   Here it is:

    I have a guest w2008R2 OS - installed in a 5.1 version of VM 7 vSpherer - which has an Emulex LPe 12002-E 8 GB dual HBA port is assigned to the as 2 PCI (1 device per port) via DirectPath devices

    The guest operating system has an application that performs backup tests.   WWPN of the HBA is zoned in the fabric and declared in the interface of Compellent Storage Center and everything is good here. the application creates a replay of a volume, mount it as a drive, read all the blocks, then dismantles and moves then repeat the process with what ever additional volumes should be safeguarded.   It is so cool I have do this with a VM right and not a physical host?  (would be cool if I had work NPIV but that's another story, not with technology HBA vendor support...)

    So, what's the problem?   Well, the bed the mounted disk backup will work extremely slow as slow as 5MB\sec flow, then during any other activity takes place on the system the flow jumps up to 150MB\sec all of a sudden and will continue to run fast for a long time and finish jobs on time with each work taking a little more than an hour.  But when the slow going the same work (all jobs are the same, they are simply by reading all the blocks of a bunch of 200 GB volumes and wrote a set of NULL value, do not forget this is just test) takes up to 10 hours to complete.

    I think that what is the culprit?   POWER MANAGEMENT.

    I have just about each performance counter and the journal of events known to man and I don't see anything abnormal on the table, ESXi virtual machine, nothing.  But what I see in the newspapers is any kind of activity that could happen in a transparent manner to save energy.

    FACTS on this case:

    -Reservation of memory is defined (it must be for directpath otherwise, you can't power on the virtual machine)

    -Booking CPU - has not been defined before an hour ago, I gave her just a small reserve for this rule, so I do not know if this will help alleviate the slow flow, or have an effect on what I think is a question of power (unit linked) management

    -windows power management is set to high performance never-ever - but not until an hour there doesn't I changed advanced settings for USB Disabled------turned off.  PCI peripheral energy saving setting was already off.  Which is in contradiction with my theory.  But maybe, just maybe, something is broken somewhere, or could be a bug... blah blah blah

    -the HBA, Emulex, as it appears in Device Manager in the guest OS doesn't have a power management tab to disable the feature, I don't see that the adapter VMXNET3 had this tab, and I turn off the power save setting to be conservative, do not think that it was the culprit, but as there is no real data running through the NIC.  The data is simply to be read from the disk mounted, not copied on any wire Ethernet.  never the less, in the Emulex properties is a Details tab and a selection of power data, and performance data indicate you what power mode, the device is, as well as feeding supported systems.  D0: full power, but D3, which shows as supported by the device, is an energy saving mode.   I could not yet attend the setting at the time of slow activity, my good intention as soon as possible. He is currently shooting in D0, full power.

    This is so what I have to offer on this, I hope that its just...   If I'm missing something, or if others have experienced this massive loss of unknown flow without reason and have resolutions, please share.

    If there is a way for me to disable power on the Emulex HBA management inform you please, I don't mind if I stop to enter the firmware, if that was even possible.  But if possible I would disable it immediately before any new tests are executed.

    The root cause of this issue has never been identified by VMware support which for me is a question of ugly.  Reservation of CPU for the VM using directpath IO with HBA installed in it corrected the problem.  I'd be leary with direct path IO.

  • [Pre-sales] WS10 ultra widescreen and Windows 8.1

    Hello

    I own WS9. I installed the latest version 9.0.3 install VM Windows 8.1 and Windows Server R2 2012.

    I have a problem with the display.

    I use 3 full screen HD related width NVidia Surround Mode with 2 graphics adapter GTX590 connected in SLI Mode.

    With all previous versions of windows and open suze, I use a 5760 x 1080 screen.

    With my Windows 2012R2 Windows 8.1 years it remains black in this resolution.

    My questions are:

    1. If I upgrade my WS 9 to 10, think you that resolution 5760 x 1080 will work?

    1. Can I test WS 10 without a license, just to check if it works?

    Thanks for your answers,


    Best regards


    Laurent

    A few suggestions:

    (1) for the change (upgrade) versions of Workstation, uninstall the version (it will ask you if you want to keep the license and the config data, say yes) and then reboot before installing the new version. Always do this when there is a change to a different version, so be it: more-> most recent or most recent-> older. I always followed it and have not had a problem installing/uninstalling (been doing this for several years to WS 5 and 6)

    (2) If you think you might want or need to go back to an older version of Workstation, using existing virtual machines, do not update/change the material of the virtual machine. If you upgrade, it may not work when go back to an older version of Workstation. (This can be corrected, but you then have to reinstall the older version of Workstation and downgrade Hardware virtual machine - just a pain to do).

    (3) when the upgrade/change the Workstation version, it is much more secure to ensure that all virtual machines are 'stop' (and not 'suspended') before doing so. Starting a VM suspended with a different version of the workstation is not a good idea. I work sometimes and not other times and I even trash a VM (I didn't a snapshot to return).

    4) "I suppose that my off VM will not be affected by this, right?" - OK, the virtual machines that are not started should not be affected by your test. (At least I've never seen a problem)

    FWIW: WS: 10.0.1 has its share of problems right now especially with the help of the invited VMTools. Just browse the forums using key words like "hgfs" and "vmtools" for more details. IMO, WS 10 fixes a lot of things, but the current version violates as a few essentials for me and as was the case with the latest major versions some, continuous WS to get a slower flow on the USB connection or network traffic with every new version major and material more fast does not seem to help at all. If it was not to support Windows 8.1, Unity Linux and USB3 Office, I would use WS 7.1.x.

    YMMV.

    Good luck.

    Post edited by: Bill Johnson, correction of typo in the version of WS

  • Windows Pro/Ent 8.1 Overview invited - stop Windows Net Performance, away from work

    I have a powerful ultimate Workstation Windows 7 x 64.  On that, I installed VMware Workstation 9.  It is perfectly up-to-date.

    Windows 8.1 Overview Pro and Windows 8.1 company Preview guests can transfer groups of files to/from the host or other virtual machines only.   The performance is dismal.

    File transfers seem to start to be able to access the files some pretty quickly, but almost immediately everything slows to a crawl, slowing down transfers down to near zero data rate.  It is trivial to reproduce - I just need to try to draw (copy) a file with a few tens of megabytes of files between host and guest.  There is almost no support network, such as measured by the the Task Manager, resource monitor, etc., but the VM and often host the two often seem bogged down.

    All the other guests Windows, including Windows XP, Windows Vista, Windows 7 and Windows 8 RTM working great - no problem.  Network transfers run as fast as the multi-plate will support.

    Networking is covered in all cases.

    Observations which may be relevant:

    • The transfer can copy multiple files quickly, then slows down almost immediately to a virtual crawl.  I've occasionally seen it fails, but usually he struggles along at speeds incredibly slow flow.
    • There is no problem to reach the Internet (e.g., navigation) of the Windows 8.1 to its customers.  It is clean and fast.  Only Windows network - for example, copy files between Windows systems - is affected.
    • The problem seems to be symmetric; for example, Windows Networking from the host to the guest, or between virtual machines Windows guest is too slow no matter who initiates the copies.
    • Sometimes late, so the client host system access, as if a common resource is tied up.  Sometimes fans will speed up, as if there is a heavy CPU load, although I've not seen much of it in the Task Manager.
    • PING the host to the host is IPv6 and takes over 100% of the time and the time < 1ms, with VERY occasional delays longer than - even if a late development is being transferred.  The problem seems to be in the network somewhere over IP stack.
    • Some file types seemed to be transferred more quickly than others, leading to wonder initially if anti-malware software (I use Avast! on the host) could be involved, but it is not consistent.  Disable the AV software does not help.  Later, I thought it was probably just a matter where the .exe and .dll files are usually larger.
    • It doesn't matter whether you use a net mapped drive or a UNC, the problem file name occurs both ways.
    • No obvious errors appear in the case of the newspapers give a clue what is happening, even if the event log is not unique (for example, "the server {AB807329-7324-431B-8B36-DBD581F56E0B} is not registered with DCOM within the required.").  However, these errors are not compatible with the operation of the network lag.
    • The problem occurs with a bone stock, just installed Windows 8.1 comments, even without the VMware tools installed.
    • Other host and VMs are working perfectly.  It is a system well maintained and perfectly stable.

    This is typical:

    SlowWin81Xfer.png

    According my control of double and triple, this isn't a simple configuration problem, although I'll be happy to check the query parameters.

    My intuition tells me the adapter virtual network is somehow incompatible with Windows 8, 1, but that's a guess.

    Thanks in advance for any help you can offer.  Please let me know that if I can provide more information, especially if there is information of diagnosis somewhere I'm neglecting.

    -Christmas

    I don't know if this can be done via the GUI.  Edit the .vmx (with the virtual machine off) and replace all occurrences of "e1000e' by 'vmxnet3.

  • Flash CS5 Preloader Questions

    Hey everybody. I'm new to the stage of Flash and I was wondering how to make a super simple preloader, which would then display my external site. I've seen many tutorials, but they never seem to work well for me. All I want to do is display text in percentage that it loads. I know that orders and the code necessary to get this number, but it never shows up when I simulate download. Where should I put the preloader and how do I do load external .swf files and redirect after loading? I can build out of a simple construction, but I need help on that part. Once I see an easy example, I should be fine. I'll add a loading bar later but I want just text at the moment. Anyway, thanks for reading.

    All help very much appreciated.

    ~ SacredDragon1337

    You can do everything correctly. Using test animation is not a good method to test the progress of a preloader, at least not without adapting it the test. Using just test movie will show usually not much or even any preload progress events. This is because the file is not really be downloaded, it is in memory and so it loads immediately.

    To get a more accurate approximation of the loading process, use test movie, then go to the view menu and select the download settings. In this list choose a slow flow, said 546 k. Then select Download simulate them. Now, you will get to see the preload progress events take place.

Maybe you are looking for

  • I forgot my password, how can I connect?

    HP PAVILION DV7WINDOWS 7Help me please! Today, I tried to turn on my laptop and it came to the normal screen where I drag my finger, so I did it and he said I had to enter a password. So I entered my password which I know is correct and it came with

  • Firefox 7.0.1 works with RoboForm?

    Firefox 7.0.1 works with RoboForm? Thank you. Lynn Spitler

  • Always get "an unknown error has occurred." When you try to reset the password.

    I got an email from iCloud and forgot his password. I'm 100% sure that safety issues are correct, but always gets "an unknown error has occurred." pop up when I press continue after having answered the questions. What answer I go, good or bad, will g

  • HARD drive randomly stops

    Then, I had this problem for about two months now, increasingly more often. My laptop (Pavilion Dv6 - 1350eb) would just freeze or stop for no apparent reason. because it was during the course of the demanding graphically programs (inventor/autocad)

  • Broken in HP Photosmart C5280 plastic gears

    I have a problem with my beloved printer) HP Photosmart C5280 broken part (10 + PC GF) and specifically small plastic gear.In this part got 3 speed, while broke which is in the middle and attached to a metal rod.Where can I find these pieces, or you