DPDK init problems

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 25, 2016, 10:25:47 AM2/25/16

to ScyllaDB users

Hello,

So I'm trying to get scylla 0.18 to run in dpdk mode on CentOS 7. And whenever I run "sudo systemctl start scylla-server" I get "map_all_hugepages(): open failed: Permission denied" indicating the something was not run in sudo.

The error trace in particular:

scylla[31163]: EAL: No free hugepages reported in hugepages-1048576kB

scylla[31163]: EAL: VFIO modules not all loaded, skip VFIO support...

scylla[31163]: EAL: Setting up physically contiguous memory...

scylla[31163]: EAL: map_all_hugepages(): open failed: Permission denied

scylla[31163]: EAL: Failed to mmap 2 MB hugepages

scylla[31163]: PANIC in rte_eal_init():

scylla[31163]: Cannot init memory

scylla[31163]: 9: [/usr/bin/scylla() [0x4c6571]]

scylla[31163]: 8: [/lib64/libc.so.6(__libc_start_main+0xf5) [0x7f485a20ab15]]

scylla[31163]: 7: [/usr/bin/scylla() [0x47043b]]

scylla[31163]: 6: [/usr/bin/scylla() [0x56a9bc]]

scylla[31163]: 5: [/usr/bin/scylla() [0x51751f]]

scylla[31163]: 4: [/usr/bin/scylla() [0x57190b]]

scylla[31163]: 3: [/usr/bin/scylla() [0x6bfb9d]]

scylla[31163]: 2: [/usr/bin/scylla() [0x469f3a]]

scylla[31163]: 1: [/usr/bin/scylla() [0x6c5b28]]

systemd[1]: scylla-server.service: main process exited, code=dumped, status=6/ABRT

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

And My /etc/sysconfig/scylla-server:

# choose following mode: virtio, dpdk, posix

NETWORK_MODE=dpdk

# tap device name(virtio)

TAP=tap0

# bridge device name (virtio)

BRIDGE=virbr0

# ethernet device name

IFNAME=$IFNAME

# setup NIC's interrupts, RPS, XPS (posix)

SET_NIC=no

# ethernet device driver (dpdk)

ETHDRV=uio_pci_generic

# ethernet device PCI ID (dpdk)

ETHPCIID=XX:XX.X

# number of hugepages

NR_HUGEPAGES=64

# user for process (must be root for dpdk)

USER=root

# group for process

GROUP=scylla

# scylla home dir

SCYLLA_HOME=/var/lib/scylla

# scylla config dir

SCYLLA_CONF=/etc/scylla

# scylla arguments (for posix mode)

#SCYLLA_ARGS="--log-to-syslog 1 --log-to-stdout 0 --default-log-level info --collectd-address=127.0.0.1:25826 --collectd=1 --collectd-poll-period 3000 --network-stack posix"

## scylla arguments (for dpdk mode)

SCYLLA_ARGS="--log-to-syslog 1 --log-to-stdout 0 -m80G --default-log-level info --collectd-address=127.0.0.1:25826 --collectd=1 --collectd-poll-period 3000 --network-stack native --dpdk-pmd"

# scylla io

SCYLLA_IO=

# setup as AMI instance

AMI=no

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 25, 2016, 10:26:39 AM2/25/16

to ScyllaDB users, chb2mn@gmail.com

Is this a red herring or how can I fix this problem?

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 25, 2016, 3:31:57 PM2/25/16

to ScyllaDB users, chb2mn@gmail.com

Solution: systemctl resets the $USER field to scylla, which I believe will overwrite the value set in /etc/systemctl/scylla-server

SYSTEMD :(

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 25, 2016, 4:10:23 PM2/25/16

to ScyllaDB users, chb2mn@gmail.com

Currently the startup process freezes  right after "Port", and I have no clue what isn't happening:

 scylla[27419]: Port 0: Enabling HW FC
 scylla[27419]: PMD: ixgbe_set_rsc(): enabling LRO mode
 scylla[27419]: PMD: ixgbe_set_rx_function(): Port[0] doesn't meet Vector Rx preconditions or RTE_IXG    BE_INC_VECTOR is not enabled
 scylla[27419]: PMD: ixgbe_set_rx_function(): LRO is requested. Using a bulk allocation version
 scylla[27419]: Checking link status
 scylla[27419]: Created DPDK device
 scylla[27419]: .....................................................................................    .....done
 scylla[27419]: Port 
[EOF]

Dor Laor

<dor@scylladb.com>

unread,

Feb 25, 2016, 4:34:09 PM2/25/16

to ScyllaDB users, chb2mn@gmail.com

Can you please strace it?

--
You received this message because you are subscribed to the Google Groups "ScyllaDB users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com.
To post to this group, send email to scyllad...@googlegroups.com.
Visit this group at https://groups.google.com/group/scylladb-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/scylladb-users/ec8c514c-c6d3-4d67-98eb-90b4b35de702%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 26, 2016, 7:29:46 PM2/26/16

to ScyllaDB users, chb2mn@gmail.com

Also, running it without systemctl has revealed that `Link Down` is printed after `Port`

$strace sudo scylla --network-stack native --dpdk-pmd

execve("/usr/bin/sudo", ["sudo", "scylla", "--network-stack", "native", "--dpdk-pmd"], [/* 27 vars */]) = 0

brk(0) = 0x7f73c859c000

fcntl(0, F_GETFD) = 0

fcntl(1, F_GETFD) = 0

fcntl(2, F_GETFD) = 0

access(0x7f73c7ee23a2, F_OK) = -1 ENOENT (No such file or directory)

mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f73c80e6000

access(0x7f73c7ee4190, R_OK) = -1 ENOENT (No such file or directory)

open(0x7f73c7ee2ab2, O_RDONLY|O_CLOEXEC) = 3

fstat(3, {...}) = 0

mmap(NULL, 36286, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7f73c80dd000

close(3) = 0

open(0x7f73c80e6640, O_RDONLY|O_CLOEXEC) = 3

read(3, 0x7ffe43ce7a30, 832) = 832

fstat(3, {...}) = 0

mmap(NULL, 2253384, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f73c7c9f000

mprotect(0x7f73c7cba000, 2097152, PROT_NONE) = 0

mmap(0x7f73c7eba000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1b000) = 0x7f73c7eba000

mmap(0x7f73c7ebc000, 37448, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7f73c7ebc000

close(3) = 0

open(0x7f73c80e6b08, O_RDONLY|O_CLOEXEC) = 3

read(3, 0x7ffe43ce7a00, 832) = 832

fstat(3, {...}) = 0

mmap(NULL, 2246784, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f73c7a7a000

mprotect(0x7f73c7a9b000, 2097152, PROT_NONE) = 0

mmap(0x7f73c7c9b000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x21000) = 0x7f73c7c9b000

mmap(0x7f73c7c9d000, 6272, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7f73c7c9d000

close(3) = 0

open(0x7f73c80e6fd0, O_RDONLY|O_CLOEXEC) = 3

read(3, 0x7ffe43ce79d0, 832) = 832

fstat(3, {...}) = 0

mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f73c80dc000

mmap(NULL, 2105616, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f73c7877000

mprotect(0x7f73c7879000, 2093056, PROT_NONE) = 0

mmap(0x7f73c7a78000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1000) = 0x7f73c7a78000

close(3) = 0

open(0x7f73c80dc4b0, O_RDONLY|O_CLOEXEC) = 3

read(3, 0x7ffe43ce79a0, 832) = 832

fstat(3, {...}) = 0

mmap(NULL, 2109744, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f73c7673000

mprotect(0x7f73c7676000, 2093056, PROT_NONE) = 0

mmap(0x7f73c7875000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x2000) = 0x7f73c7875000

close(3) = 0

open(0x7f73c80dc978, O_RDONLY|O_CLOEXEC) = 3

read(3, 0x7ffe43ce7970, 832) = 832

fstat(3, {...}) = 0

mmap(NULL, 3932736, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f73c72b2000

mprotect(0x7f73c7468000, 2097152, PROT_NONE) = 0

mmap(0x7f73c7668000, 24576, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1b6000) = 0x7f73c7668000

mmap(0x7f73c766e000, 16960, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7f73c766e000

close(3) = 0

open(0x7f73c80dce68, O_RDONLY|O_CLOEXEC) = 3

read(3, 0x7ffe43ce7720, 832) = 832

fstat(3, {...}) = 0

mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f73c80db000

mmap(NULL, 2490888, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f73c7051000

mprotect(0x7f73c70b1000, 2093056, PROT_NONE) = 0

mmap(0x7f73c72b0000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x5f000) = 0x7f73c72b0000

close(3) = 0

open(0x7f73c80db4b0, O_RDONLY|O_CLOEXEC) = 3

read(3, 0x7ffe43ce76f0, 832) = 832

fstat(3, {...}) = 0

mmap(NULL, 2245240, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f73c6e2c000

mprotect(0x7f73c6e50000, 2093056, PROT_NONE) = 0

mmap(0x7f73c704f000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x23000) = 0x7f73c704f000

close(3) = 0

open(0x7f73c80dba68, O_RDONLY|O_CLOEXEC) = 3

read(3, 0x7ffe43ce7690, 832) = 832

fstat(3, {...}) = 0

mmap(NULL, 2208864, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f73c6c10000

mprotect(0x7f73c6c26000, 2097152, PROT_NONE) = 0

mmap(0x7f73c6e26000, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x16000) = 0x7f73c6e26000

mmap(0x7f73c6e28000, 13408, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7f73c6e28000

close(3) = 0

mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f73c80da000

mmap(NULL, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f73c80d8000

arch_prctl(ARCH_SET_FS, 0x7f73c80d8800) = 0

mprotect(0x7f73c7668000, 16384, PROT_READ) = 0

mprotect(0x7f73c6e26000, 4096, PROT_READ) = 0

mprotect(0x7f73c704f000, 4096, PROT_READ) = 0

mprotect(0x7f73c72b0000, 4096, PROT_READ) = 0

mprotect(0x7f73c7875000, 4096, PROT_READ) = 0

mprotect(0x7f73c7a78000, 4096, PROT_READ) = 0

mprotect(0x7f73c7c9b000, 4096, PROT_READ) = 0

mprotect(0x7f73c7eba000, 4096, PROT_READ) = 0

mprotect(0x7f73c8307000, 4096, PROT_READ) = 0

mprotect(0x7f73c80e7000, 4096, PROT_READ) = 0

munmap(0x7f73c80dd000, 36286) = 0

set_tid_address(0x7f73c80d8ad0) = 15349

set_robust_list(0x7f73c80d8ae0, 24) = 0

rt_sigaction(SIGRTMIN, {...}, NULL, 8) = 0

rt_sigaction(SIGRT_1, {...}, NULL, 8) = 0

rt_sigprocmask(SIG_UNBLOCK, 0x7ffe43ce80f8, NULL, 8) = 0

getrlimit(RLIMIT_STACK, 0x7ffe43ce80e0) = 0

statfs(0x7f73c7a93c6a, {...}) = 0

statfs(0x7f73c7a93c71, 0x7ffe43ce80c0) = -1 ENOENT (No such file or directory)

brk(0) = 0x7f73c859c000

brk(0x7f73c85bd000) = 0x7f73c85bd000

open(0x7f73c7a93c4e, O_RDONLY) = 3

fstat(3, {...}) = 0

mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f73c80e5000

read(3, 0x7f73c80e5000, 1024) = 310

stat(0x7f73c7430e08, 0x7ffe43ce7fa0) = -1 ENOENT (No such file or directory)

close(3) = 0

munmap(0x7f73c80e5000, 4096) = 0

open(0x7f73c7a93c7a, O_RDONLY) = 3

fstat(3, {...}) = 0

mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f73c80e5000

read(3, 0x7f73c80e5000, 1024) = 754

read(3, "", 1024) = 0

close(3) = 0

munmap(0x7f73c80e5000, 4096) = 0

open(0x7f73c7431af0, O_RDONLY|O_CLOEXEC) = 3

fstat(3, {...}) = 0

mmap(NULL, 106065056, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7f73c06e9000

close(3) = 0

geteuid() = 1421200090

open(0x7ffe43ce7ad0, O_RDONLY|O_CLOEXEC) = 3

fstat(3, {...}) = 0

mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f73c80e5000

read(3, 0x7f73c80e5000, 4096) = 2502

read(3, "", 4096) = 0

close(3) = 0

munmap(0x7f73c80e5000, 4096) = 0

open(0x7f73c859d100, O_RDONLY) = -1 ENOENT (No such file or directory)

open(0x7f73c859e190, O_RDONLY) = -1 ENOENT (No such file or directory)

open(0x7f73c859d190, O_RDONLY) = -1 ENOENT (No such file or directory)

open(0x7f73c859e110, O_RDONLY) = -1 ENOENT (No such file or directory)

open(0x7f73c859e220, O_RDONLY) = -1 ENOENT (No such file or directory)

open(0x7f73c859d210, O_RDONLY) = -1 ENOENT (No such file or directory)

write(2, 0x7ffe43ce96e8, 4sudo) = 4

write(2, 0x7f73c8101897, 2: ) = 2

write(2, 0x7ffe43ce5810, 54effective uid is not 0, is sudo installed setuid root?) = 54

write(2, 0x7f73c766d243, 1

) = 1

close(0) = 0

access(0x7f73c742fd27, F_OK) = -1 ENOENT (No such file or directory)

open(0x7f73c742fd36, O_RDONLY|O_CLOEXEC) = 0

lseek(0, 0, SEEK_SET) = 0

alarm(0) = 0

rt_sigaction(SIGALRM, {...}, {...}, 8) = 0

alarm(10) = 0

fcntl(0, F_SETLKW, {...}) = 0

read(0, 0x7f73c76719a0, 384) = 384

read(0, "", 384) = 0

fcntl(0, F_SETLKW, {...}) = 0

alarm(0) = 10

rt_sigaction(SIGALRM, {...}, NULL, 8) = 0

exit_group(1) = ?

+++ exited with 1 +++

Avi Kivity

<avi@scylladb.com>

unread,

Feb 28, 2016, 3:54:02 AM2/28/16

to scylladb-users@googlegroups.com, chb2mn@gmail.com

On 02/27/2016 02:29 AM, chb...@gmail.com wrote:

Also, running it without systemctl has revealed that `Link Down` is printed after `Port`

Is that port connected?

$strace sudo scylla --network-stack native --dpdk-pmd

Use 'sudo strace -fF' instead of 'strace sudo' to get file names printed out correctly (and -fF for threads).

scylla[31163]: 3: [/usr/bin/scy lla() [0x6bfb9d]]

-- You received this message because you are subscribed to the Google Groups "ScyllaDB users" group. To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com. To post to this group, send email to scyllad...@googlegroups.com. Visit this group at https://groups.google.com/group/scylladb-users. To view this discussion on the web visit https://groups.google.com/d/msgid/scylladb-users/a49cdca3-d2b0-44d8-a62a-ddf9a9ddc254%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 28, 2016, 3:14:00 PM2/28/16

to ScyllaDB users, chb2mn@gmail.com

> Is that port connected?

Apparently not? It's strange though because I can bind that port to an IP and ssh into there and things work just fine.

I changed NICs entirely and it worked and my throughput was 1000MBps which IIRC is suboptimal, right?

Avi Kivity

<avi@scylladb.com>

unread,

Feb 29, 2016, 4:08:41 AM2/29/16

to scylladb-users@googlegroups.com, chb2mn@gmail.com

On 02/28/2016 10:14 PM, chb...@gmail.com wrote:

> Is that port connected?

Apparently not? It's strange though because I can bind that port to an IP and ssh into there and things work just fine.

Maybe the cable or port was flakey so it took more time to negotiate than the dpdk timeout.

I changed NICs entirely and it worked and my throughput was 1000MBps which IIRC is suboptimal, right?

1000MBps = 10,000 Mbps, which is pretty good.

-- You received this message because you are subscribed to the Google Groups "ScyllaDB users" group. To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com. To post to this group, send email to scyllad...@googlegroups.com. Visit this group at https://groups.google.com/group/scylladb-users. To view this discussion on the web visit https://groups.google.com/d/msgid/scylladb-users/e45807f2-50c1-432d-8bf2-39702c27fab3%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 29, 2016, 11:30:36 AM2/29/16

to ScyllaDB users, chb2mn@gmail.com

> Maybe the cable or port was flakey so it took more time to negotiate than the dpdk timeout.

Perhaps, I'll investigate. Most likely, those NICs were put there on accident and never fully connected.

> 1000MBps = 10,000 Mbps, which is pretty good.

1,000 Mbps*, my mistake

Avi Kivity

<avi@scylladb.com>

unread,

Feb 29, 2016, 11:33:40 AM2/29/16

to scylladb-users@googlegroups.com, chb2mn@gmail.com

On 02/29/2016 06:30 PM, chb...@gmail.com wrote:

> Maybe the cable or port was flakey so it took more time to negotiate than the dpdk timeout.

Perhaps, I'll investigate. Most likely, those NICs were put there on accident and never fully connected.

> 1000MBps = 10,000 Mbps, which is pretty good.

1,000 Mbps*, my mistake

I see. How did you measure this?

-- You received this message because you are subscribed to the Google Groups "ScyllaDB users" group. To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com. To post to this group, send email to scyllad...@googlegroups.com. Visit this group at https://groups.google.com/group/scylladb-users. To view this discussion on the web visit https://groups.google.com/d/msgid/scylladb-users/501d8a14-6d81-4e32-addd-4d9e3fd3f32e%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 29, 2016, 12:18:12 PM2/29/16

to ScyllaDB users, chb2mn@gmail.com

It's the output when I switched NICs. It worked and started gossiping with other nodes.

Checking link status

Created DPDK device

.................................done

Port 0 Link Up - speed 1000 Mbps - full-duplex

DHCP sending discover

DHCP timeout

Scylla version 0.18-20160216.b532919 starting ...

...

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 29, 2016, 12:36:08 PM2/29/16

to ScyllaDB users, chb2mn@gmail.com

I now get a series of ARP timeouts from the nodes in the cluster. Potentially due to the connection not being fast enough?

INFO [shard 0] rpc - client $MY_CLIENT: client connection dropped: ARP timeout

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Feb 29, 2016, 3:37:17 PM2/29/16

to ScyllaDB users, chb2mn@gmail.com

FYI $MY_CLIENT is the other seed in the cluster. So essentially, they can't talk to each other because (I believe) the timeout is too short and my connection is not fast enough.

When trying to use the Ethernet Controller 10-Gigabit X540-AT2, I get the Link Down error.

When using the I350 Gigabit Network Connection I get a a 1,000 Mb link which is too slow

Dor Laor

<dor@scylladb.com>

unread,

Feb 29, 2016, 3:43:51 PM2/29/16

to ScyllaDB users, Christopher Burkhalter

On Mon, Feb 29, 2016 at 10:37 PM, <chb...@gmail.com> wrote:

FYI $MY_CLIENT is the other seed in the cluster. So essentially, they can't talk to each other because (I believe) the timeout is too short and my connection is not fast enough.

When trying to use the Ethernet Controller 10-Gigabit X540-AT2, I get the Link Down error.

It's probably a pure dpdk driver issue.

I'd check the most uptodate dpdk version and use their test app and report back to us.

We may need a rebase

To view this discussion on the web visit https://groups.google.com/d/msgid/scylladb-users/0d814826-c224-4cd6-b30f-bcf6d6827fc8%40googlegroups.com.

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Mar 1, 2016, 9:29:50 AM3/1/16

to ScyllaDB users, chb2mn@gmail.com

I did using DPDK v2.2.0 and I was able to run their example helloworld program using those interfaces.

Tutorial for which is found here: http://dpdk.org/doc/guides/sample_app_ug/hello_world.html

chb2mn@gmail.com

<chb2mn@gmail.com>

unread,

Mar 1, 2016, 11:39:15 AM3/1/16

to ScyllaDB users, chb2mn@gmail.com

I've seen here that scylla uses dpdk version 1.8? Is this still true? will Scylla rebase before GA or should I downgrade dpdk?

...

Avi Kivity

<avi@scylladb.com>

unread,

Mar 1, 2016, 11:44:00 AM3/1/16

to scylladb-users@googlegroups.com, chb2mn@gmail.com

We're using 2.1.0 (+ local patches), and likely will only rebase after GA.

-- You received this message because you are subscribed to the Google Groups "ScyllaDB users" group. To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com. To post to this group, send email to scyllad...@googlegroups.com. Visit this group at https://groups.google.com/group/scylladb-users. To view this discussion on the web visit https://groups.google.com/d/msgid/scylladb-users/335730b0-d8a4-4e33-a7d9-7a08780e30c7%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.

Reply all

Reply to author

Forward